text
stringlengths
29
3.76k
label
stringclasses
8 values
We propose a novel approach to designing RDF-stores with the goal of improving the consistency and predictability of query performance. When designing these systems, three properties are commonly desired: support for the full range of SPARQL query features (Q), support for widely varying RDF datasets in terms of structuredness and size (S), and high performance (P). We develop the empirical SPQ conjecture which states that it may be impossible to achieve all the three desiderata simultaneously. We present a strong case for its plausibility based on our experimental results. The tradeoffs among the three and design guidelines based on the SPQ conjecture are also discussed.
cs.*
Computation is classically studied in terms of automata, formal languages and algorithms; yet, the relation between neural dynamics and symbolic representations and operations is still unclear in traditional eliminative connectionism. Therefore, we suggest a unique perspective on this central issue, to which we would like to refer as to transparent connectionism, by proposing accounts of how symbolic computation can be implemented in neural substrates. In this study we first introduce a new model of dynamics on a symbolic space, the versatile shift, showing that it supports the real-time simulation of a range of automata. We then show that the Goedelization of versatile shifts defines nonlinear dynamical automata, dynamical systems evolving on a vectorial space. Finally, we present a mapping between nonlinear dynamical automata and recurrent artificial neural networks. The mapping defines an architecture characterized by its granular modularity, where data, symbolic operations and their control are not only distinguishable in activation space, but also spatially localizable in the network itself, while maintaining a distributed encoding of symbolic representations. The resulting networks simulate automata in real-time and are programmed directly, in absence of network training. To discuss the unique characteristics of the architecture and their consequences, we present two examples: i) the design of a Central Pattern Generator from a finite-state locomotive controller, and ii) the creation of a network simulating a system of interactive automata that supports the parsing of garden-path sentences as investigated in psycholinguistics experiments.
cs.*
We develop methods for accelerating metric similarity search that are effective on modern hardware. Our algorithms factor into easily parallelizable components, making them simple to deploy and efficient on multicore CPUs and GPUs. Despite the simple structure of our algorithms, their search performance is provably sublinear in the size of the database, with a factor dependent only on its intrinsic dimensionality. We demonstrate that our methods provide substantial speedups on a range of datasets and hardware platforms. In particular, we present results on a 48-core server machine, on graphics hardware, and on a multicore desktop.
cs.*
Groundwater is one of the most vital of all common pool resources throughout the world. More than half of groundwater is used to grow crops. This research models groundwater depletion patterns within a multi-agent system framework. Irrigators are modeled as agents in the multi-agent system. The irrigation strategies adopted by the agents are investigated using game theory. A set of five irrigators, growing three crops: corn, sorghum and wheat, have been considered in this study. To allow groundwater flow, these agents are assumed to be located in adjoining farm lands. Irrigators are modeled selfish agents that strategize their irrigation patterns in order to maximize their own utilities, i.e. the difference between the total revenue obtained from crop sales and the costs incurred, including groundwater extraction costs. Due to groundwater flow, and have no incentive to conserve groundwater. This leads to unsustainable depletion of the resource under Nash equilibrium, when no irrigator can increase its utility by unilaterally changing its strategy. All parameters in this research are representative of Kansas. Recorded environmental and economic data of the region, along with the DSSAT software, have been used to obtain these futuristic projections. One of the emergent phenomena of the simulations is the adoption of crop rotation patterns by the irrigators to conserve groundwater. The irrigators grow corn, which is a more profitable yet water intensive crop in one year, and in the next, conserve water by growing sorghum instead. Another emergent outcome of this research is the viability of LEMAs. When the irrigators are subject to LEMA-level limits on groundwater use, there is a slight increase in the aggregate utility of the LEMA.
cs.*
Improving the effectiveness and safety of patient care is the ultimate objective for medical cyber-physical systems. Many medical best practice guidelines exist, but most of the existing guidelines in handbooks are difficult for medical staff to remember and apply clinically. Furthermore, although the guidelines have gone through clinical validations, validations by medical professionals alone do not provide guarantees for the safety of medical cyber-physical systems. Hence, formal verification is also needed. The paper presents the formal semantics for a framework that we developed to support the development of verifiably safe medical guidelines. The framework allows computer scientists to work together with medical professionals to transform medical best practice guidelines into executable statechart models, Yakindu in particular, so that medical functionalities and properties can be quickly prototyped and validated. Existing formal verification technologies, UPPAAL timed automata in particular, is integrated into the framework to provide formal verification capabilities to verify safety properties. However, some components used/built into the framework, such as the open-source Yakindu statecharts as well as the transformation rules from statecharts to timed automata, do not have built-in semantics. The ambiguity becomes unavoidable unless formal semantics is defined for the framework, which is what the paper is to present.
cs.*
Automated Market Makers (AMMs) are decentralized applications that allow users to exchange crypto-tokens without the need for a matching exchange order. AMMs are one of the most successful DeFi use cases: indeed, major AMM platforms process a daily volume of transactions worth USD billions. Despite their popularity, AMMs are well-known to suffer from transaction-ordering issues: adversaries can influence the ordering of user transactions, and possibly front-run them with their own, to extract value from AMMs, to the detriment of users. We devise an effective procedure to construct a strategy through which an adversary can maximize the value extracted from user transactions.
cs.*
Contemporary use of the term 'intension' derives from the traditional logical Frege-Russell's doctrine that an idea (logic formula) has both an extension and an intension. From the Montague's point of view, the meaning of an idea can be considered as particular extensions in different possible worlds. In this paper we analyze the minimal intensional semantic enrichment of the syntax of the FOL language, by unification of different views: Tarskian extensional semantics of the FOL, modal interpretation of quantifiers, and a derivation of the Tarskian theory of truth from unified semantic theory based on a single meaning relation. We show that not all modal predicate logics are intensional, and that an equivalent modal Kripke's interpretation of logic quantifiers in FOL results in a particular pure extensional modal predicate logic (as is the standard Tarskian semantics of the FOL). This minimal intensional enrichment is obtained by adopting the theory of properties, relations and propositions (PRP) as the universe or domain of the FOL, composed by particulars and universals (or concepts), with the two-step interpretation of the FOL that eliminates the weak points of the Montague's intensional semantics. Differently from the Bealer's intensional FOL, we show that it is not necessary the introduction of the intensional abstraction in order to obtain the full intensional properties of the FOL. Final result of this paper is represented by the commutative homomorphic diagram that holds in each given possible world of this new intensional FOL, from the free algebra of the FOL syntax, toward its intensional algebra of concepts, and, successively, to the new extensional relational algebra (different from Cylindric algebras), and we show that it corresponds to the Tarski's interpretation of the standard extensional FOL in this possible world.
cs.*
Understanding the relationship between the composition of a research team and the potential impact of their research papers is crucial as it can steer the development of new science policies for improving the research enterprise. Numerous studies assess how the characteristics and diversity of research teams can influence their performance across several dimensions: ethnicity, internationality, size, and others. In this paper, we explore the impact of diversity in terms of the authors' expertise. To this purpose, we retrieved 114K papers in the field of Computer Science and analysed how the diversity of research fields within a research team relates to the number of citations their papers received in the upcoming 5 years. The results show that two different metrics we defined, reflecting the diversity of expertise, are significantly associated with the number of citations. This suggests that, at least in Computer Science, diversity of expertise is key to scientific impact.
cs.*
We adapt an existing preemptive scheduling model of RTOS kernel by eChronos from machine-assisted proof to Spin-based model checker. The model we constructed can be automatically verified rather than formulating proofs by hand. Moreover, we look into the designs of a Linux-like real-time kernel--Piko/RT and the specification of ARMv7-M architecture to reconstruct the model, and use LTL to specify a simple concurrent programs--consumer/producer problem during the development stage of the kernel. We show that under the preemptive scheduling and the mechanism of ARMv7-M, the program will not suffer from race condition, starvation, and deadlock.
cs.*
Logic Programming languages and combinational circuit synthesis tools share a common "combinatorial search over logic formulae" background. This paper attempts to reconnect the two fields with a fresh look at Prolog encodings for the combinatorial objects involved in circuit synthesis. While benefiting from Prolog's fast unification algorithm and built-in backtracking mechanism, efficiency of our search algorithm is ensured by using parallel bitstring operations together with logic variable equality propagation, as a mapping mechanism from primary inputs to the leaves of candidate Leaf-DAGs implementing a combinational circuit specification. After an exhaustive expressiveness comparison of various minimal libraries, a surprising first-runner, Strict Boolean Inequality "<" together with constant function "1" also turns out to have small transistor-count implementations, competitive to NAND-only or NOR-only libraries. As a practical outcome, a more realistic circuit synthesizer is implemented that combines rewriting-based simplification of (<,1) circuits with exhaustive Leaf-DAG circuit search. Keywords: logic programming and circuit design, combinatorial object generation, exact combinational circuit synthesis, universal boolean logic libraries, symbolic rewriting, minimal transistor-count circuit synthesis
cs.*
Numerous microarchitectural optimizations unlocked tremendous processing power for deep neural networks that in turn fueled the AI revolution. With the exhaustion of such optimizations, the growth of modern AI is now gated by the performance of training systems, especially their data movement. Instead of focusing on single accelerators, we investigate data-movement characteristics of large-scale training at full system scale. Based on our workload analysis, we design HammingMesh, a novel network topology that provides high bandwidth at low cost with high job scheduling flexibility. Specifically, HammingMesh can support full bandwidth and isolation to deep learning training jobs with two dimensions of parallelism. Furthermore, it also supports high global bandwidth for generic traffic. Thus, HammingMesh will power future large-scale deep learning systems with extreme bandwidth requirements.
cs.*
This technical report covers a set of experiments on the 64-core SPARC T3-4 system, comparing it to two similar AMD and Intel systems. Key characteristics as maximum integer and floating point arithmetic throughput are measured as well as memory throughput, showing the scalability of the SPARC T3-4 system. The performance of POSIX threads primitives is characterized and compared in detail, such as thread creation and mutex synchronization. Scalability tests with a fine grained multithreaded runtime are performed, showing problems with atomic CAS operations on such physically highly parallel systems.
cs.*
Virtualization technologies have evolved along with the development of computational environments since virtualization offered needed features at that time such as isolation, accountability, resource allocation, resource fair sharing and so on. Novel processor technologies bring to commodity computers the possibility to emulate diverse environments where a wide range of computational scenarios can be run. Along with processors evolution, system developers have created different virtualization mechanisms where each new development enhanced the performance of previous virtualized environments. Recently, operating system-based virtualization technologies captured the attention of communities abroad (from industry to academy and research) because their important improvements on performance area. In this paper, the features of three container-based operating systems virtualization tools (LXC, Docker and Singularity) are presented. LXC, Docker, Singularity and bare metal are put under test through a customized single node HPL-Benchmark and a MPI-based application for the multi node testbed. Also the disk I/O performance, Memory (RAM) performance, Network bandwidth and GPU performance are tested for the COS technologies vs bare metal. Preliminary results and conclusions around them are presented and discussed.
cs.*
Basic mirroring (BM) classified as RAID level 1 replicates data on two disks, thus doubling disk access bandwidth for read requests. RAID1/0 is an array of BM pairs with balanced loads due to striping. When a disk fails the read load on its pair is doubled, which results in halving the maximum attainable bandwidth. We review RAID1 organizations which attain a balanced load upon disk failure, but as shown by reliability analysis tend to be less reliable than RAID1/0. Hybrid disk arrays which store XORed instead of replicated data tend to have a higher reliability than mirrored disks, but incur a higher overhead in updating data. Read request response time can be improved by processing them at a higher priority than writes, since they have a direct effect on application response time. Shortest seek distance and affinity based routing both shorten seek time. Anticipatory arm placement places arms optimally to minimize the seek distance. The analysis of RAID1 in normal, degraded, and rebuild mode is provided to quantify RAID1/0 performance. We compare the reliability of mirrored disk organizations against each other and hybrid disks and erasure coded disk arrays.
cs.*
Cloud providers offer a variety of execution platforms in form of bare-metal, VM, and containers. However, due to the pros and cons of each execution platform, choosing the appropriate platform for a specific cloud-based application has become a challenge for solution architects. The possibility to combine these platforms (e.g. deploying containers within VMs) offers new capacities that makes the challenge even further complicated. However, there is a little study in the literature on the pros and cons of deploying different application types on various execution platforms. In particular, evaluation of diverse hardware configurations and different CPU provisioning methods, such as CPU pinning, have not been sufficiently studied in the literature. In this work, the performance overhead of container, VM, and bare-metal execution platforms are measured and analyzed for four categories of real-world applications, namely video processing, parallel processing (MPI), web processing, and No-SQL, respectively representing CPU intensive, parallel processing, and two IO intensive processes. Our analyses reveal a set of interesting and sometimes counterintuitive findings that can be used as best practices by the solution architects to efficiently deploy cloud-based applications. Here are some notable mentions: (A) Under specific circumstances, containers can impose a higher overhead than VMs; (B) Containers on top of VMs can mitigate the overhead of VMs for certain applications; (C) Containers with a large number of cores impose a lower overhead than those with a few cores.
cs.*
Modern external memory is represented by several device classes. At present, HDD, SATA SSD and NVMe SSD are widely used. Recently ultra-low latency SSD such as Intel Optane became available on the market. Each of these types exhibits it's own pattern for throughput, latency and parallelism. To achieve the highest performance one has to pick an appropriate I/O interface provided by the operating system. In this work we present a detailed overview and evaluation of modern storage reading performance with regard to available Linux synchronous and asynchronous interfaces. While throughout this work we aim for the highest throughput we also measure latency and CPU usage. We provide this report in hope the detailed results could be interesting to both researchers and practitioners.
cs.*
The serverless computing model strengthens the cloud computing tendency to abstract resource management. Serverless platforms are responsible for deploying and scaling the developer's applications. Serverless also incorporated the pay-as-you-go billing model, which only considers the time spent processing client requests. Such a decision created a natural incentive for improving the platform's efficient resource usage. This search for efficiency can lead to the cold start problem, which represents a delay to execute serverless applications. Among the solutions proposed to deal with the cold start, those based on the snapshot method stand out. Despite the rich exploration of the technique, there is a lack of research that evaluates the solution's trade-offs. In this direction, this work compares two solutions to mitigate the cold start: Prebaking and SEUSS. We analyzed the solution's performance with functions of different levels of complexity: NoOp, a function that renders Markdown to HTML, and a function that loads 41 MB of dependencies. Preliminary results indicated that Prebaking showed a 33% and 25% superior performance to startup the NoOp and Markdown functions, respectively. Further analysis also revealed that Prebaking's warmup mechanism reduced the Markdown first request processing time by 69%.
cs.*
The growing pressure on cloud application scalability has accentuated storage performance as a critical bottle- neck. Although cache replacement algorithms have been extensively studied, cache prefetching - reducing latency by retrieving items before they are actually requested remains an underexplored area. Existing approaches to history-based prefetching, in particular, provide too few benefits for real systems for the resources they cost. We propose MITHRIL, a prefetching layer that efficiently exploits historical patterns in cache request associations. MITHRIL is inspired by sporadic association rule mining and only relies on the timestamps of requests. Through evaluation of 135 block-storage traces, we show that MITHRIL is effective, giving an average of a 55% hit ratio increase over LRU and PROBABILITY GRAPH, a 36% hit ratio gain over AMP at reasonable cost. We further show that MITHRIL can supplement any cache replacement algorithm and be readily integrated into existing systems. Furthermore, we demonstrate the improvement comes from MITHRIL being able to capture mid-frequency blocks.
cs.*
Robustness of a distributed computing system is defined as the ability to maintain its performance in the presence of uncertain parameters. Uncertainty is a key problem in heterogeneous (and even homogeneous) distributed computing systems that perturbs system robustness. Notably, the performance of these systems is perturbed by uncertainty in both task execution time and arrival. Accordingly, our goal is to make the system robust against these uncertainties. Considering task execution time as a random variable, we use probabilistic analysis to develop an autonomous proactive task dropping mechanism to attain our robustness goal. Specifically, we provide a mathematical model that identifies the optimality of a task dropping decision, so that the system robustness is maximized. Then, we leverage the mathematical model to develop a task dropping heuristic that achieves the system robustness within a feasible time complexity. Although the proposed model is generic and can be applied to any distributed system, we concentrate on heterogeneous computing (HC) systems that have a higher degree of exposure to uncertainty than homogeneous systems. Experimental results demonstrate that the autonomous proactive dropping mechanism can improve the system robustness by up to 20%.
cs.*
Most modern operating systems have adopted the one-to-one thread model to support fast execution of threads in both multi-core and single-core systems. This thread model, which maps the kernel-space and user-space threads in a one-to-one manner, supports quick thread creation and termination in high-performance server environments. However, the performance of time-critical threads is degraded when multiple threads are being run in low-end CE devices with limited system resources. When a CE device runs many threads to support diverse application functionalities, low-level hardware specifications often lead to significant resource contention among the threads trying to obtain system resources. As a result, the operating system encounters challenges, such as excessive thread context switching overhead, execution delay of time-critical threads, and a lack of virtual memory for thread stacks. This paper proposes a state-of-the-art Thread Evolution Kit (TEK) that consists of three primary components: a CPU Mediator, Stack Tuner, and Enhanced Thread Identifier. From the experiment, we can see that the proposed scheme significantly improves user responsiveness (7x faster) under high CPU contention compared to the traditional thread model. Also, TEK solves the segmentation fault problem that frequently occurs when a CE application increases the number of threads during its execution.
cs.*
Modern enterprise servers are increasingly embracing tiered memory systems with a combination of low latency DRAMs and large capacity but high latency non-volatile main memories (NVMMs) such as Intel's Optane DC PMM. Prior works have focused on efficient placement and migration of data on a tiered memory system, but have not studied the optimal placement of page tables. Explicit and efficient placement of page tables is crucial for large memory footprint applications with high TLB miss rates because they incur dramatically higher page walk latency when page table pages are placed in NVMM. We show that (i) page table pages can end up on NVMM even when enough DRAM memory is available and (ii) page table pages that spill over to NVMM due to DRAM memory pressure are not migrated back later when memory is available in DRAM. We study the performance impact of page table placement in a tiered memory system and propose an efficient and transparent page table management technique that (i) applies different placement policies for data and page table pages, (ii) introduces a differentiating policy for page table pages by placing a small but critical part of the page table in DRAM, and (iii) dynamically and judiciously manages the rest of the page table by transparently migrating the page table pages between DRAM and NVMM. Our implementation on a real system equipped with Intel's Optane NVMM running Linux reduces the page table walk cycles by 12% and total cycles by 20% on an average. This improves the runtime by 20% on an average for a set of synthetic and real-world large memory footprint applications when compared with various default Linux kernel techniques.
cs.*
In this paper, we propose the first optimum process scheduling algorithm for an increasingly prevalent type of heterogeneous multicore (HEMC) system that combines high-performance big cores and energy-efficient small cores with the same instruction-set architecture (ISA). Existing algorithms are all heuristics-based, and the well-known IPC-driven approach essentially tries to schedule high scaling factor processes on big cores. Our analysis shows that, for optimum solutions, it is also critical to consider placing long running processes on big cores. Tests of SPEC 2006 cases on various big-small core combinations show that our proposed optimum approach is up to 34% faster than the IPC-driven heuristic approach in terms of total workload completion time. The complexity of our algorithm is O(NlogN) where N is the number of processes. Therefore, the proposed optimum algorithm is practical for use.
cs.*
Modern I/O applications that run on HPC infrastructures are increasingly becoming read and metadata intensive. However, having multiple concurrent applications submitting large amounts of metadata operations can easily saturate the shared parallel file system's metadata resources, leading to overall performance degradation and I/O unfairness. We present PADLL, an application and file system agnostic storage middleware that enables QoS control of data and metadata workflows in HPC storage systems. It adopts ideas from Software-Defined Storage, building data plane stages that mediate and rate limit POSIX requests submitted to the shared file system, and a control plane that holistically coordinates how all I/O workflows are handled. We demonstrate its performance and feasibility under multiple QoS policies using synthetic benchmarks, real-world applications, and traces collected from a production file system. Results show that PADLL can enforce complex storage QoS policies over concurrent metadata-aggressive jobs, ensuring fairness and prioritization.
cs.*
We present DIO, a generic tool for observing inefficient and erroneous I/O interactions between applications and in-kernel storage systems that lead to performance, dependability, and correctness issues. DIO facilitates the analysis and enables near real-time visualization of complex I/O patterns for data-intensive applications generating millions of storage requests. This is achieved by non-intrusively intercepting system calls, enriching collected data with relevant context, and providing timely analysis and visualization for traced events. We demonstrate its usefulness by analyzing two production-level applications. Results show that DIO enables diagnosing resource contention in multi-threaded I/O that leads to high tail latency and erroneous file accesses that cause data loss.
cs.*
The performance of data intensive applications is often dominated by their input/output (I/O) operations but the I/O stack of systems is complex and severely depends on system specific settings and hardware components. This situation makes generic performance optimisation challenging and costly for developers as they would have to run their application on a large variety of systems to evaluate their improvements. Here, simulation frameworks can help reducing the experimental overhead but they typically handle the topic of I/O rather coarse-grained, which leads to significant inaccuracies in performance predictions. Here, we propose a more accurate model of the write performance of Linux-based systems that takes different I/O methods and levels (via system calls, library calls, direct or indirect, etc.), the page cache, background writing, and the I/O throttling capabilities of the Linux kernel into account. With our model, we reduce, for example, the relative prediction error compared to a standard I/O model included in SimGrid for a random I/O scenario from 67 % down to 10 % relative error against real measurements of the simulated workload. In other scenarios the differences are even more pronounced.
cs.*
RAID proposal advocated replacing large disks with arrays of PC disks, but as the capacity of small disks increased 100-fold in 1990s the production of large disks was discontinued. Storage dependability is increased via replication or erasure coding. Cloud storage providers store multiple copies of data obviating for need for further redundancy. Varitaions of RAID based on local recovery codes, partial MDS reduce recovery cost. NAND flash Solid State Disks - SSDs have low latency and high bandwidth, are more reliable, consume less power and have a lower TCO than Hard Disk Drives, which are more viable for hyperscalers.
cs.*
A classical result in descriptive complexity theory states that Datalog expresses exactly the class of polynomially computable queries on ordered databases. In this paper we extend this result to the case of higher-order Datalog. In particular, we demonstrate that on ordered databases, for all $k\geq2$, $k$-order Datalog captures $(k-1)$-EXPTIME. This result suggests that higher-order extensions of Datalog possess superior expressive power and they are worthwhile of further investigation both in theory and in practice. This paper is under consideration for acceptance in TPLP.
cs.*
In the field of robust geometric computation it is often necessary to make exact decisions based on inexact floating-point arithmetic. One common approach is to store the computation history in an arithmetic expression dag and to re-evaluate the expression with increasing precision until an exact decision can be made. We show that exact-decisions number types based on expression dags can be evaluated faster in practice through parallelization on multiple cores. We compare the impact of several restructuring methods for the expression dag on its running time in a parallel environment.
cs.*
We present algorithmic, complexity and implementation results concerning real root isolation of integer univariate polynomials using the continued fraction expansion of real algebraic numbers. One motivation is to explain the method's good performance in practice. We improve the previously known bound by a factor of $d \tau$, where $d$ is the polynomial degree and $\tau$ bounds the coefficient bitsize, thus matching the current record complexity for real root isolation by exact methods. Namely, the complexity bound is $\sOB(d^4 \tau^2)$ using the standard bound on the expected bitsize of the integers in the continued fraction expansion. We show how to compute the multiplicities within the same complexity and extend the algorithm to non square-free polynomials. Finally, we present an efficient open-source \texttt{C++} implementation in the algebraic library \synaps, and illustrate its efficiency as compared to other available software. We use polynomials with coefficient bitsize up to 8000 and degree up to 1000.
cs.*
Recent technological advances are enabling HCI researchers to explore interaction possibilities for remote XR collaboration using high-fidelity reconstructions of physical activity spaces. However, creating these reconstructions often lacks user involvement with an overt focus on capturing sensory context that does not necessarily augment an informal social experience. This work seeks to understand social context that can be important for reconstruction to enable XR applications for informal instructional scenarios. Our study involved the evaluation of an XR remote guidance prototype by 8 intergenerational groups of closely related gardeners using reconstructions of personally meaningful spaces in their gardens. Our findings contextualize physical objects and areas with various motivations related to gardening and detail perceptions of XR that might affect the use of reconstructions for remote interaction. We discuss implications for user involvement to create reconstructions that better translate real-world experience, encourage reflection, incorporate privacy considerations, and preserve shared experiences with XR as a medium for informal intergenerational activities.
cs.*
This major component of the research described in this thesis is 3D computer graphics, specifically the realistic physics-based softbody simulation and haptic responsive environments. Minor components include advanced human-computer interaction environments, non-linear documentary storytelling, and theatre performance. The journey of this research has been unusual because it requires a researcher with solid knowledge and background in multiple disciplines; who also has to be creative and sensitive in order to combine the possible areas into a new research direction. [...] It focuses on the advanced computer graphics and emerges from experimental cinematic works and theatrical artistic practices. Some development content and installations are completed to prove and evaluate the described concepts and to be convincing. [...] To summarize, the resulting work involves not only artistic creativity, but solving or combining technological hurdles in motion tracking, pattern recognition, force feedback control, etc., with the available documentary footage on film, video, or images, and text via a variety of devices [....] and programming, and installing all the needed interfaces such that it all works in real-time. Thus, the contribution to the knowledge advancement is in solving these interfacing problems and the real-time aspects of the interaction that have uses in film industry, fashion industry, new age interactive theatre, computer games, and web-based technologies and services for entertainment and education. It also includes building up on this experience to integrate Kinect- and haptic-based interaction, artistic scenery rendering, and other forms of control. This research work connects all the research disciplines, seemingly disjoint fields of research, such as computer graphics, documentary film, interactive media, and theatre performance together.
cs.*
Implementing new, high-performance MAC protocols requires real-time features, to be able to synchronize correctly between different unrelated devices. Such features are highly desirable for operating wireless sensor networks (WSN) that are designed to be part of the Internet of Things (IoT). Unfortunately, the operating systems commonly used in this domain cannot provide such features. On the other hand, "bare-metal" development sacrifices portability, as well as the mul-titasking abilities needed to develop the rich applications that are useful in the domain of the Internet of Things. We describe in this paper how we helped solving these issues by contributing to the development of a port of RIOT OS on the MSP430 microcontroller, an architecture widely used in IoT-enabled motes. RIOT OS offers rich and advanced real-time features, especially the simultaneous use of as many hardware timers as the underlying platform (microcontroller) can offer. We then demonstrate the effectiveness of these features by presenting a new implementation, on RIOT OS, of S-CoSenS, an efficient MAC protocol that uses very low processing power and energy.
cs.*
In recent years, more and more applications use ad-hoc networks for local M2M communications, but in some cases such as when using WSNs, the software processing delay induced by packets relaying may not be negligible. In this paper, we planned and carried out a delay measurement experiment using Raspberry Pi Zero W. The results demonstrated that, in low-energy ad-hoc networks, processing delay of the application is always too large to ignore; it is at least ten times greater than the kernel routing and corresponds to 30% of the transmission delay. Furthermore, if the task is CPU-intensive, such as packet encryption, the processing delay can be greater than the transmission delay and its behavior is represented by a simple linear model. Our findings indicate that the key factor for achieving QoS in ad-hoc networks is an appropriate node-to-node load balancing that takes into account the CPU performance and the amount of traffic passing through each node.
cs.*
Wireless Sensor Networks (WSNs) are being used in various applications such as structural health monitoring and industrial control. Since energy efficiency is one of the major design factors, the existing WSNs primarily rely on low-power, low-rate wireless technologies such as 802.15.4 and Bluetooth. In this paper, we strive to tackle the challenges of developing ultra-high-rate WSNs based on 802.11 (WiFi) standard by proposing Sensifi. As an illustrative application of this system, we consider vibration test monitoring of spacecraft and identify system design requirements and challenges. Our main contributions are as follows. First, we propose packet encoding methods to reduce the overhead of assigning accurate timestamps to samples. Second, we propose energy efficiency methods to enhance the system's lifetime. Third, we reduce the overhead of processing outgoing packets through network stack to enhance sampling rate and mitigate sampling rate instability. Fourth, we study and reduce the delay of processing incoming packets through network stack to enhance the accuracy of time synchronization among nodes. Fifth, we propose a low-power node design for ultra-high-rate applications. Sixth, we use our node design to empirically evaluate the system.
cs.*
We provide linear-time algorithms for geometric graphs with sublinearly many crossings. That is, we provide algorithms running in O(n) time on connected geometric graphs having n vertices and k crossings, where k is smaller than n by an iterated logarithmic factor. Specific problems we study include Voronoi diagrams and single-source shortest paths. Our algorithms all run in linear time in the standard comparison-based computational model; hence, we make no assumptions about the distribution or bit complexities of edge weights, nor do we utilize unusual bit-level operations on memory words. Instead, our algorithms are based on a planarization method that "zeroes in" on edge crossings, together with methods for extending planar separator decompositions to geometric graphs with sublinearly many crossings. Incidentally, our planarization algorithm also solves an open computational geometry problem of Chazelle for triangulating a self-intersecting polygonal chain having n segments and k crossings in linear time, for the case when k is sublinear in n by an iterated logarithmic factor.
cs.*
Inspired by distributed applications that use consensus or other agreement protocols for global coordination, we define a new computational model for parameterized systems that is based on a general global synchronization primitive and allows for global transition guards. Our model generalizes many existing models in the literature, including broadcast protocols and guarded protocols. We show that reachability properties are decidable for systems without guards, and give sufficient conditions under which they remain decidable in the presence of guards. Furthermore, we investigate cutoffs for reachability properties and provide sufficient conditions for small cutoffs in a number of cases that are inspired by our target applications.
cs.*
We introduce a formal framework for analyzing trades in financial markets. An exchange is where multiple buyers and sellers participate to trade. These days, all big exchanges use computer algorithms that implement double sided auctions to match buy and sell requests and these algorithms must abide by certain regulatory guidelines. For example, market regulators enforce that a matching produced by exchanges should be \emph{fair}, \emph{uniform} and \emph{individual rational}. To verify these properties of trades, we first formally define these notions in a theorem prover and then give formal proofs of relevant results on matchings. Finally, we use this framework to verify properties of two important classes of double sided auctions. All the definitions and results presented in this paper are completely formalised in the Coq proof assistant without adding any additional axioms to it.
cs.*
We use the incompressibility method based on Kolmogorov complexity to determine the total number of bits of routing information for almost all network topologies. In most models for routing, for almost all labeled graphs $\Theta (n^2)$ bits are necessary and sufficient for shortest path routing. By `almost all graphs' we mean the Kolmogorov random graphs which constitute a fraction of $1-1/n^c$ of all graphs on $n$ nodes, where $c > 0$ is an arbitrary fixed constant. There is a model for which the average case lower bound rises to $\Omega(n^2 \log n)$ and another model where the average case upper bound drops to $O(n \log^2 n)$. This clearly exposes the sensitivity of such bounds to the model under consideration. If paths have to be short, but need not be shortest (if the stretch factor may be larger than 1), then much less space is needed on average, even in the more demanding models. Full-information routing requires $\Theta (n^3)$ bits on average. For worst-case static networks we prove a $\Omega(n^2 \log n)$ lower bound for shortest path routing and all stretch factors $<2$ in some networks where free relabeling is not allowed.
cs.*
A data stream model represents setting where approximating pairwise, or $k$-wise, independence with sublinear memory is of considerable importance. In the streaming model the joint distribution is given by a stream of $k$-tuples, with the goal of testing correlations among the components measured over the entire stream. In the streaming model, Indyk and McGregor (SODA 08) recently gave exciting new results for measuring pairwise independence. The Indyk and McGregor methods provide $\log{n}$-approximation under statistical distance between the joint and product distributions in the streaming model. Indyk and McGregor leave, as their main open question, the problem of improving their $\log n$-approximation for the statistical distance metric. In this paper we solve the main open problem posed by of Indyk and McGregor for the statistical distance for pairwise independence and extend this result to any constant $k$. In particular, we present an algorithm that computes an $(\epsilon, \delta)$-approximation of the statistical distance between the joint and product distributions defined by a stream of $k$-tuples. Our algorithm requires $O(({1\over \epsilon}\log({nm\over \delta}))^{(30+k)^k})$ memory and a single pass over the data stream.
cs.*
The pigeonhole principle states that if $n$ items are contained in $m$ boxes, then at least one box has no more than $n / m$ items. It is utilized to solve many data management problems, especially for thresholded similarity searches. Despite many pigeonhole principle-based solutions proposed in the last few decades, the condition stated by the principle is weak. It only constrains the number of items in a single box. By organizing the boxes in a ring, we propose a new principle, called the pigeonring principle, which constrains the number of items in multiple boxes and yields stronger conditions. To utilize the new principle, we focus on problems defined in the form of identifying data objects whose similarities or distances to the query is constrained by a threshold. Many solutions to these problems utilize the pigeonhole principle to find candidates that satisfy a filtering condition. By the new principle, stronger filtering conditions can be established. We show that the pigeonhole principle is a special case of the new principle. This suggests that all the pigeonhole principle-based solutions are possible to be accelerated by the new principle. A universal filtering framework is introduced to encompass the solutions to these problems based on the new principle. Besides, we discuss how to quickly find candidates specified by the new principle. The implementation requires only minor modifications on top of existing pigeonhole principle-based algorithms. Experimental results on real datasets demonstrate the applicability of the new principle as well as the superior performance of the algorithms based on the new principle.
cs.*
Exact Maximum Inner Product Search (MIPS) is an important task that is widely pertinent to recommender systems and high-dimensional similarity search. The brute-force approach to solving exact MIPS is computationally expensive, thus spurring recent development of novel indexes and pruning techniques for this task. In this paper, we show that a hardware-efficient brute-force approach, blocked matrix multiply (BMM), can outperform the state-of-the-art MIPS solvers by over an order of magnitude, for some -- but not all -- inputs. In this paper, we also present a novel MIPS solution, MAXIMUS, that takes advantage of hardware efficiency and pruning of the search space. Like BMM, MAXIMUS is faster than other solvers by up to an order of magnitude, but again only for some inputs. Since no single solution offers the best runtime performance for all inputs, we introduce a new data-dependent optimizer, OPTIMUS, that selects online with minimal overhead the best MIPS solver for a given input. Together, OPTIMUS and MAXIMUS outperform state-of-the-art MIPS solvers by 3.2$\times$ on average, and up to 10.9$\times$, on widely studied MIPS datasets.
cs.*
We propose and analyze a new Markov Chain Monte Carlo algorithm that generates a uniform sample over full and non-full dimensional polytopes. This algorithm, termed "Matrix Hit and Run" (MHAR), is a modification of the Hit and Run framework. For the regime $n^{1+\frac{1}{3}} \ll m$, MHAR has a lower asymptotic cost per sample in terms of soft-O notation ($\SO$) than do existing sampling algorithms after a \textit{warm start}. MHAR is designed to take advantage of matrix multiplication routines that require less computational and memory resources. Our tests show this implementation to be substantially faster than the \textit{hitandrun} R package, especially for higher dimensions. Finally, we provide a python library based on Pytorch and a Colab notebook with the implementation ready for deployment in architectures with GPU or just CPU.
cs.*
In this paper, a methodology is presented and employed for simulating the Internet of Things (IoT). The requirement for scalability, due to the possibly huge amount of involved sensors and devices, and the heterogeneous scenarios that might occur, impose resorting to sophisticated modeling and simulation techniques. In particular, multi-level simulation is regarded as a main framework that allows simulating large-scale IoT environments while keeping high levels of detail, when it is needed. We consider a use case based on the deployment of smart services in decentralized territories. A two level simulator is employed, which is based on a coarse agent-based, adaptive parallel and distributed simulation approach to model the general life of simulated entities. However, when needed a finer grained simulator (based on OMNeT++) is triggered on a restricted portion of the simulated area, which allows considering all issues concerned with wireless communications. Based on this use case, it is confirmed that the ad-hoc wireless networking technologies do represent a principle tool to deploy smart services over decentralized countrysides. Moreover, the performance evaluation confirms the viability of utilizing multi-level simulation for simulating large scale IoT environments.
cs.*
This paper presents a methodology for simulating the Internet of Things (IoT) using multi-level simulation models. With respect to conventional simulators, this approach allows us to tune the level of detail of different parts of the model without compromising the scalability of the simulation. As a use case, we have developed a two-level simulator to study the deployment of smart services over rural territories. The higher level is base on a coarse grained, agent-based adaptive parallel and distributed simulator. When needed, this simulator spawns OMNeT++ model instances to evaluate in more detail the issues concerned with wireless communications in restricted areas of the simulated world. The performance evaluation confirms the viability of multi-level simulations for IoT environments.
cs.*
Large-scale decentralized systems of autonomous agents interacting via asynchronous communication often experience the following self-healing dilemma: fault detection inherits network uncertainties making a remote faulty process indistinguishable from a slow process. In the case of a slow process without fault, fault correction is undesirable as it can trigger new faults that could be prevented with fault tolerance that is a more proactive system maintenance. But in the case of an actual faulty process, fault tolerance alone without eventually correcting persistent faults can make systems underperforming. Measuring, understanding and resolving such self-healing dilemmas is a timely challenge and critical requirement given the rise of distributed ledgers, edge computing, the Internet of Things in several energy, transport and health applications. This paper contributes a novel and general-purpose modeling of fault scenarios during system runtime. They are used to accurately measure and predict inconsistencies generated by the undesirable outcomes of fault correction and fault tolerance as the means to improve self-healing of large-scale decentralized systems at the design phase. A rigorous experimental methodology is designed that evaluates 696 experimental settings of different fault scales, fault profiles and fault detection thresholds in a prototyped decentralized network of 3000 nodes. Almost 9 million measurements of inconsistencies were collected in a network, where each node monitors the health status of another node, while both can defect. The prediction performance of the modeled fault scenarios is validated in a challenging application scenario of decentralized and dynamic in-network data aggregation using real-world data from a Smart Grid pilot project. Findings confirm the origin of inconsistencies at design phase.
cs.*
This paper deals with the problem of properly simulating the Internet of Things (IoT). Simulating an IoT allows evaluating strategies that can be employed to deploy smart services over different kinds of territories. However, the heterogeneity of scenarios seriously complicates this task. This imposes the use of sophisticated modeling and simulation techniques. We discuss novel approaches for the provision of scalable simulation scenarios, that enable the real-time execution of massively populated IoT environments. Attention is given to novel hybrid and multi-level simulation techniques that, when combined with agent-based, adaptive Parallel and Distributed Simulation (PADS) approaches, can provide means to perform highly detailed simulations on demand. To support this claim, we detail a use case concerned with the simulation of vehicular transportation systems.
cs.*
Multiuser resource allocation has recently been recognized as an effective methodology for enhancing the power and spectrum efficiency in OFDM (orthogonal frequency division multiplexing) systems. It is, however, not directly applicable to current packet-switched networks, because (i) most existing packet-scheduling schemes are based on a single-server model and do not serve multiple users at the same time; and (ii) the conventional separate design of MAC (medium access control) packet scheduling and PHY (physical) resource allocation yields inefficient resource utilization. In this paper, we propose a cross-layer resource allocation algorithm based on a novel multi-server scheduling framework to achieve overall high system power efficiency in packet-switched OFDM networks. Our contribution is four fold: (i) we propose and analyze a MPGPS (multi-server packetized general processor sharing) service discipline that serves multiple users at the same time and facilitates multiuser resource allocation; (ii) we present a MPGPS-based joint MAC-PHY resource allocation scheme that incorporates packet scheduling, subcarrier allocation, and power allocation in an integrated framework; (iii) by investigating the fundamental tradeoff between multiuser-diversity and queueing performance, we present an A-MPGPS (adaptive MPGPS) service discipline that strikes balance between power efficiency and queueing performance; and (iv) we extend MPGPS to an O-MPGPS (opportunistic MPGPS) service discipline to further enhance the resource utilization efficiency.
cs.*
The upcoming many-core architectures require software developers to exploit concurrency to utilize available computational power. Today's high-level language virtual machines (VMs), which are a cornerstone of software development, do not provide sufficient abstraction for concurrency concepts. We analyze concrete and abstract concurrency models and identify the challenges they impose for VMs. To provide sufficient concurrency support in VMs, we propose to integrate concurrency operations into VM instruction sets. Since there will always be VMs optimized for special purposes, our goal is to develop a methodology to design instruction sets with concurrency support. Therefore, we also propose a list of trade-offs that have to be investigated to advise the design of such instruction sets. As a first experiment, we implemented one instruction set extension for shared memory and one for non-shared memory concurrency. From our experimental results, we derived a list of requirements for a full-grown experimental environment for further research.
cs.*
Thirty-six years after the first edition of IEEE standard 982.1, Measures of the Software Aspects of Dependability, the third edition focuses on the measurement of in-service software dependability. This article explains how this new point of view evolved and shaped the third edition's guidance for software dependability measurement.
cs.*
A Frontal-Delaunay refinement algorithm for mesh generation in piecewise smooth domains is described. Built using a restricted Delaunay framework, this new algorithm combines a number of novel features, including: (i) an unweighted, conforming restricted Delaunay representation for domains specified as a (non-manifold) collection of piecewise smooth surface patches and curve segments, (ii) a protection strategy for domains containing curve segments that subtend sharply acute angles, and (iii) a new class of off-centre refinement rules designed to achieve high-quality point-placement along embedded curve features. Experimental comparisons show that the new Frontal-Delaunay algorithm outperforms a classical (statically weighted) restricted Delaunay-refinement technique for a number of three-dimensional benchmark problems.
cs.*
Compute Express Link (CXL) is a rapidly emerging coherent interconnect standard that provides opportunities for memory pooling and sharing. Memory sharing is a well-established software feature that improves memory utilization by avoiding unnecessary data movement. In this paper, we discuss multiple approaches to enable memory sharing with different generations of CXL protocol (i.e., CXL 2.0 and CXL 3.0) considering the challenges with each of the architectures from the device hardware and software viewpoint.
cs.*
Memory accounts for 33 - 50% of the total cost of ownership (TCO) in modern data centers. We propose a novel solution to tame memory TCO through the novel creation and judicious management of multiple software-defined compressed memory tiers. As opposed to the state-of-the-art solutions that employ a 2-Tier solution, a single compressed tier along with DRAM, we define multiple compressed tiers implemented through a combination of different compression algorithms, memory allocators for compressed objects, and backing media to store compressed objects. These compressed memory tiers represent distinct points in the access latency, data compressibility, and unit memory usage cost spectrum, allowing rich and flexible trade-offs between memory TCO savings and application performance impact. A key advantage with ntier is that it enables aggressive memory TCO saving opportunities by placing warm data in low latency compressed tiers with a reasonable performance impact while simultaneously placing cold data in the best memory TCO saving tiers. We believe our work represents an important server system configuration and optimization capability to achieve the best SLA-aware performance per dollar for applications hosted in production data center environments. We present a comprehensive and rigorous analytical cost model for performance and TCO trade-off based on continuous monitoring of the application's data access profile. Guided by this model, our placement model takes informed actions to dynamically manage the placement and migration of application data across multiple software-defined compressed tiers. On real-world benchmarks, our solution increases memory TCO savings by 22% - 40% percentage points while maintaining performance parity or improves performance by 2% - 10% percentage points while maintaining memory TCO parity compared to state-of-the-art 2-Tier solutions.
cs.*
ADG is a forum to exchange ideas and views, to present research results and progress, and to demonstrate software tools at the intersection between geometry and automated deduction. The conference is held every two years. The previous editions of ADG were held in Hagenberg in 2021 (online, postponed from 2020 due to COVID-19), Nanning in 2018, Strasbourg in 2016, Coimbra in 2014, Edinburgh in 2012, Munich in 2010, Shanghai in 2008, Pontevedra in 2006, Gainesville in 2004, Hagenberg in 2002, Zurich in 2000, Beijing in 1998, and Toulouse in 1996. The 14th edition, ADG 2023, was held in Belgrade, Serbia, in September 20-22, 2023. This edition of ADG had an additional special focus topic, Deduction in Education. Invited Speakers: Julien Narboux, University of Strasbourg, France "Formalisation, arithmetization and automatisation of geometry"; Filip Mari\'c, University of Belgrade, Serbia, "Automatization, formalization and visualization of hyperbolic geometry"; Zlatan Magajna, University of Ljubljana, Slovenia, "Workshop OK Geometry"
cs.*
The development of scalable, representative, and widely adopted benchmarks for graph data systems have been a question for which answers has been sought for decades. We conduct an in-depth study of the existing literature on benchmarks for graph data management and processing, covering 20 different benchmarks developed during the last 15 years. We categorize the benchmarks into three areas focusing on benchmarks for graph processing systems, graph database benchmarks, and bigdata benchmarks with graph processing workloads. This systematic approach allows us to identify multiple issues existing in this area, including i) few benchmarks exist which can produce high workload scenarios, ii) no significant work done on benchmarking graph stream processing as well as graph based machine learning, iii) benchmarks tend to use conventional metrics despite new meaningful metrics have been around for years, iv) increasing number of big data benchmarks appear with graph processing workloads. Following these observations, we conclude the survey by describing key challenges for future research on graph data systems benchmarking.
cs.*
As one of open-source codes widely used in computational ocean acoustics, FOR3D can provide a very good estimate for underwater acoustic propagation. In this paper, we propose a performance optimization and parallelization to speed up the running of FOR3D. We utilized a variety of methods to enhance the entire performance, such as using a multi-threaded programming model to exploit the potential capability of the many-core node of high-performance computing (HPC) system, tuning compile options, using efficient tuned mathematical library and utilizing vectorization optimization instruction. In addition, we extended the application from single-frequency calculation to multi-frequency calculation successfully by using OpenMP+MPI hybrid programming techniques on the mainstream HPC platform. A detailed performance evaluation was performed and the results showed that the proposed parallelization obtained good accelerated effect of 25.77X when testing a typical three-dimensional medium-sized case on Tianhe-2 supercomputer. It also showed that the tuned parallel version has a weak-scalability. The speed of calculation of underwater sound field can be greatly improved by the strategy mentioned in this paper. The method used in this paper is not only applicable to other similar computing models in computational ocean acoustics but also a guideline of performance enhancement for scientific and engineering application running on modern many-core-computing platform.
cs.*
The numerical solution of partial differential equations using the finite element method is one of the key applications of high performance computing. Local assembly is its characteristic operation. This entails the execution of a problem-specific kernel to numerically evaluate an integral for each element in the discretized problem domain. Since the domain size can be huge, executing efficient kernels is fundamental. Their op- timization is, however, a challenging issue. Even though affine loop nests are generally present, the short trip counts and the complexity of mathematical expressions make it hard to determine a single or unique sequence of successful transformations. Therefore, we present the design and systematic evaluation of COF- FEE, a domain-specific compiler for local assembly kernels. COFFEE manipulates abstract syntax trees generated from a high-level domain-specific language for PDEs by introducing domain-aware composable optimizations aimed at improving instruction-level parallelism, especially SIMD vectorization, and register locality. It then generates C code including vector intrinsics. Experiments using a range of finite-element forms of increasing complexity show that significant performance improvement is achieved.
cs.*
Algorithmic and architecture-oriented optimizations are essential for achieving performance worthy of anticipated energy-austere exascale systems. In this paper, we present an extreme scale FMM-accelerated boundary integral equation solver for wave scattering, which uses FMM as a matrix-vector multiplication inside the GMRES iterative method. Our FMM Helmholtz kernels treat nontrivial singular and near-field integration points. We implement highly optimized kernels for both shared and distributed memory, targeting emerging Intel extreme performance HPC architectures. We extract the potential thread- and data-level parallelism of the key Helmholtz kernels of FMM. Our application code is well optimized to exploit the AVX-512 SIMD units of Intel Skylake and Knights Landing architectures. We provide different performance models for tuning the task-based tree traversal implementation of FMM, and develop optimal architecture-specific and algorithm aware partitioning, load balancing, and communication reducing mechanisms to scale up to 6,144 compute nodes of a Cray XC40 with 196,608 hardware cores. With shared memory optimizations, we achieve roughly 77% of peak single precision floating point performance of a 56-core Skylake processor, and on average 60% of peak single precision floating point performance of a 72-core KNL. These numbers represent nearly 5.4x and 10x speedup on Skylake and KNL, respectively, compared to the baseline scalar code. With distributed memory optimizations, on the other hand, we report near-optimal efficiency in the weak scalability study with respect to both the logarithmic communication complexity as well as the theoretical scaling complexity of FMM. In addition, we exhibit up to 85% efficiency in strong scaling. We compute in excess of 2 billion DoF on the full-scale of the Cray XC40 supercomputer.
cs.*
With data durability, high access speed, low power efficiency and byte addressability, NVMe and SSD, which are acknowledged representatives of emerging storage technologies, have been applied broadly in many areas. However, one key issue with high-performance adoption of these technologies is how to properly define intelligent cache layers such that the performance gap between emerging technologies and main memory can be well bridged. To this end, we propose Phoebe, a reuse-aware reinforcement learning framework for the optimal online caching that is applicable for a wide range of emerging storage models. By continuous interacting with the cache environment and the data stream, Phoebe is capable to extract critical temporal data dependency and relative positional information from a single trace, becoming ever smarter over time. To reduce training overhead during online learning, we utilize periodical training to amortize costs. Phoebe is evaluated on a set of Microsoft cloud storage workloads. Experiment results show that Phoebe is able to close the gap of cache miss rate from LRU and a state-of-the-art online learning based cache policy to the Belady's optimal policy by 70.3% and 52.6%, respectively.
cs.*
We study a generalization of the classic paging problem that allows the amount of available memory to vary over time - capturing a fundamental property of many modern computing realities, from cloud computing to multi-core and energy-optimized processors. It turns out that good performance in the "classic" case provides no performance guarantees when memory capacity fluctuates: roughly speaking, moving from static to dynamic capacity can mean the difference between optimality within a factor 2 in space and time, and suboptimality by an arbitrarily large factor. More precisely, adopting the competitive analysis framework, we show that some online paging algorithms, despite having an optimal (h,k)-competitive ratio when capacity remains constant, are not (3,k)-competitive for any arbitrarily large k in the presence of minimal capacity fluctuations. In this light it is surprising that several classic paging algorithms perform remarkably well even if memory capacity changes adversarially - even without taking those changes into explicit account! In particular, we prove that LFD still achieves the minimum number of faults, and that several classic online algorithms such as LRU have a "dynamic" (h,k)-competitive ratio that is the best one can achieve without knowledge of future page requests, even if one had perfect knowledge of future capacity fluctuations (an exact characterization of this ratio shows it is almost, albeit not quite, equal to the "classic" ratio k/(k-h+1)). In other words, with careful management, knowing/predicting future memory resources appears far less crucial to performance than knowing/predicting future data accesses.
cs.*
Applications making excessive use of single-object based data structures (such as linked lists, trees, etc...) can see a drop in efficiency over a period of time due to the randomization of nodes in memory. This slow down is due to the ineffective use of the CPU's L1/L2 cache. We present a novel approach for mitigating this by presenting the design of a single-object memory allocator that preserves memory locality across randomly ordered memory allocations and deallocations.
cs.*
Applications in the AI and HPC fields require much memory capacity, and the amount of energy consumed by main memory of server machines is ever increasing. Energy consumption of main memory can be greatly reduced by applying approximate computing in exchange for increased bit error rates. AI and HPC applications are to some extent robust to bit errors because small numerical errors are amortized by their iterative nature. However, a single occurrence of a NaN due to bit-flips corrupts the whole calculation result. The issue is that fixing every bit-flip using ECC incurs too much overhead because the bit error rate is much higher than in normal environments. We propose a low-overhead method to fix NaNs when approximate computing is applied to main memory. The main idea is to reactively repair NaNs while leaving other non-fatal numerical errors as-is to reduce the overhead. We implemented a prototype by leveraging floating-point exceptions of x86 CPUs, and the preliminary evaluations showed that our method incurs negligible overhead.
cs.*
Power ISA(TM) Version 3.1 has introduced a new family of matrix math instructions, collectively known as the Matrix-Multiply Assist (MMA) facility. The instructions in this facility implement numerical linear algebra operations on small matrices and are meant to accelerate computation-intensive kernels, such as matrix multiplication, convolution and discrete Fourier transform. These instructions have led to a power- and area-efficient implementation of a high throughput math engine in the future POWER10 processor. Performance per core is 4 times better, at constant frequency, than the previous generation POWER9 processor. We also advocate the use of compiler built-ins as the preferred way of leveraging these instructions, which we illustrate through case studies covering matrix multiplication and convolution.
cs.*
Because of the increasing demand for computation in DNN, researchers develope both hardware and software mechanisms to reduce the compute and memory burden. A widely adopted approach is to use mixed precision data types. However, it is hard to leverage mixed precision without hardware support because of the overhead of data casting. Hardware vendors offer tensorized instructions for mixed-precision tensor operations, like Intel VNNI, Tensor Core, and ARM-DOT. These instructions involve a computing idiom that reduces multiple low precision elements into one high precision element. The lack of compilation techniques for this makes it hard to utilize these instructions: Using vendor-provided libraries for computationally-intensive kernels is inflexible and prevents further optimizations, and manually writing hardware intrinsics is error-prone and difficult for programmers. Some prior works address this problem by creating compilers for each instruction. This requires excessive effort when it comes to many tensorized instructions. In this work, we develop a compiler framework to unify the compilation for these instructions -- a unified semantics abstraction eases the integration of new instructions, and reuses the analysis and transformations. Tensorized instructions from different platforms can be compiled via UNIT with moderate effort for favorable performance. Given a tensorized instruction and a tensor operation, UNIT automatically detects the applicability, transforms the loop organization of the operation,and rewrites the loop body to leverage the tensorized instruction. According to our evaluation, UNIT can target various mainstream hardware platforms. The generated end-to-end inference model achieves 1.3x speedup over Intel oneDNN on an x86 CPU, 1.75x speedup over Nvidia cuDNN on an NvidiaGPU, and 1.13x speedup over a carefully tuned TVM solution for ARM DOT on an ARM CPU.
cs.*
We present the MEMA framework for the easy and quick derivation of efficient inference runtimes that minimize external memory accesses for matrix multiplication on TinyML systems. The framework accounts for hardware resource constraints and problem sizes in analytically determining optimized schedules and kernels that minimize memory accesses. MEMA provides a solution to a well-known problem in the current practice, that is, optimal schedules tend to be found only through a time consuming and heuristic search of a large scheduling space. We compare the performance of runtimes derived from MEMA to existing state-of-the-art libraries on ARM-based TinyML systems. For example, for neural network benchmarks on the ARM Cortex-M4, we achieve up to a 1.8x speedup and 44% energy reduction over CMSIS-NN.
cs.*
We propose Rosko -- row skipping outer products -- for deriving sparse matrix multiplication (SpMM) kernels in reducing computation and memory access requirements of deep neural networks (DNNs). Rosko allows skipping of entire row computations during program execution with low sparsity-management overheads. We analytically derive sparse CPU kernels that adapt to given hardware characteristics to effectively utilize processor cores and minimize data movement without the need for auto-tuning or search space exploration. Rosko can be integrated with other outer product scheduling methods, allowing them to leverage row skipping by using Rosko's packing format to skip unnecessary computation. Rosko kernels outperform existing auto-tuning and search-based solutions as well as state-of-the-art vendor-optimized libraries on real hardware across a variety of neural network workloads. For matrices with sparsities ranging from 65% to 99.8% typically found in machine learning, Rosko kernels achieve up to a 6.5x runtime reduction on Intel and ARM CPUs.
cs.*
The proliferation of open knowledge graphs has led to a surge in scholarly research on the topic over the past decade. This paper presents a bibliometric analysis of the scholarly literature on open knowledge graphs published between 2013 and 2023. The study aims to identify the trends, patterns, and impact of research in this field, as well as the key topics and research questions that have emerged. The work uses bibliometric techniques to analyze a sample of 4445 scholarly articles retrieved from Scopus. The findings reveal an ever-increasing number of publications on open knowledge graphs published every year, particularly in developed countries (+50 per year). These outputs are published in highly-referred scholarly journals and conferences. The study identifies three main research themes: (1) knowledge graph construction and enrichment, (2) evaluation and reuse, and (3) fusion of knowledge graphs into NLP systems. Within these themes, the study identifies specific tasks that have received considerable attention, including entity linking, knowledge graph embedding, and graph neural networks.
cs.*
The recursive method for computing the generalized LM-inverse of a constant rectangular matrix augmented by a column vector is proposed in Udwadia and Phohomsiri (2007) [16] and [17]. The corresponding algorithm for the sequential determination of the generalized LM-inverse is established in the present paper. We prove that the introduced algorithm for computing the generalized LM-inverse and the algorithm for the computation of the weighted Moore-Penrose inverse developed by Wang and Chen (1986) in [23] are equivalent algorithms. Both of the algorithms are implemented in the present paper using the package MATHEMATICA. Several rational test matrices and randomly generated constant matrices are tested and the CPU time is compared and discussed.
cs.*
The aim of the INFINITY workshop is to provide a forum for researchers interested in the development of formal methods and algorithmic techniques for the analysis of systems with infinitely many states, and their application in automated verification of complex software and hardware systems.
cs.*
The abstract mathematical theory of partial differential equations (PDEs) is formulated in terms of manifolds, scalar fields, tensors, and the like, but these algebraic structures are hardly recognizable in actual PDE solvers. The general aim of the Sophus programming style is to bridge the gap between theory and practice in the domain of PDE solvers. Its main ingredients are a library of abstract datatypes corresponding to the algebraic structures used in the mathematical theory and an algebraic expression style similar to the expression style used in the mathematical theory. Because of its emphasis on abstract datatypes, Sophus is most naturally combined with object-oriented languages or other languages supporting abstract datatypes. The resulting source code patterns are beyond the scope of current compiler optimizations, but are sufficiently specific for a dedicated source-to-source optimizer. The limited, domain-specific, character of Sophus is the key to success here. This kind of optimization has been tested on computationally intensive Sophus style code with promising results. The general approach may be useful for other styles and in other application domains as well.
cs.*
General Matrix Multiplication or GEMM kernels take centre place in high performance computing and machine learning. Recent NVIDIA GPUs include GEMM accelerators, such as NVIDIA's Tensor Cores. Their exploitation is hampered by the two-language problem: it requires either low-level programming which implies low programmer productivity or using libraries that only offer a limited set of components. Because rephrasing algorithms in terms of established components often introduces overhead, the libraries' lack of flexibility limits the freedom to explore new algorithms. Researchers using GEMMs can hence not enjoy programming productivity, high performance, and research flexibility at once. In this paper we solve this problem. We present three sets of abstractions and interfaces to program GEMMs within the scientific Julia programming language. The interfaces and abstractions are co-designed for researchers' needs and Julia's features to achieve sufficient separation of concerns and flexibility to easily extend basic GEMMs in many different ways without paying a performance price. Comparing our GEMMs to state-of-the-art libraries cuBLAS and CUTLASS, we demonstrate that our performance is in the same ballpark of the libraries, and in some cases even exceeds it, without having to write a single line of code in CUDA C++ or assembly, and without facing flexibility limitations.
cs.*
Over recent years heterogeneous systems have become more prevalent across HPC systems, with over 100 supercomputers in the TOP500 incorporating GPUs or other accelerators. These hardware platforms have different performance characteristics and optimization requirements. In order to make the most of multiple accelerators a developer has to provide implementations of their algorithms tuned for each device. Hardware vendors provide libraries targeting their devices specifically, which provide good performance but frequently have different API designs, hampering portability. The SYCL programming model allows users to write heterogeneous programs using completely standard C++, and so developers have access to the power of C++ templates when developing compute kernels. In this paper we show that by writing highly parameterized kernels for matrix multiplies and convolutions we achieve performance competitive with vendor implementations across different architectures. Furthermore, tuning for new devices amounts to choosing the combinations of kernel parameters that perform best on the hardware.
cs.*
In this paper we use attribute grammars as a formal approach for model checkers development. Our aim is to design an ATL (Alternating-Time Temporal Logic) model checker from a context-free grammar which generates the language of the ATL formulas. An attribute grammar may be informally defined as a context-free grammar which is extended with a set of attributes and a collection of semantic rules. We use an ATL attribute grammar for specifying an operational semantics of the language of the ATL formulas by defining a translation into the language which describes the set of states from the ATL model where the corresponding ATL formulas are satisfied. We provide a formal definition for an attribute grammar used as input for Another Tool for Language Recognition (ANTLR) to generate an ATL model checker. Also, the technique of implementing the semantic actions in ANTLR is presented, which is the concept of connection between attribute evaluation in the grammar that generates the language of ATL formulas and algebraic compiler implementation that represents the ATL model checker. The original implementation of the model checking algorithm is based on Relational Databases and Web Services. Several database systems and Web Services technologies were used for evaluating the system performance in verification of large ATL models.
cs.*
Graph layouts are key to exploring massive graphs. An enormous number of nodes and edges do not allow network analysis software to produce meaningful visualization of the pervasive networks. Long computation time, memory and display limitations encircle the software's ability to explore massive graphs. This paper introduces BigGraphVis, a new parallel graph visualization method that uses GPU parallel processing and community detection algorithm to visualize graph communities. We combine parallelized streaming community detection algorithm and probabilistic data structure to leverage parallel processing of Graphics Processing Unit (GPU). To the best of our knowledge, this is the first attempt to combine the power of streaming algorithms coupled with GPU computing to tackle big graph visualization challenges. Our method extracts community information in a few passes on the edge list, and renders the community structures using the ForceAtlas2 algorithm. Our experiment with massive real-life graphs indicates that about 70 to 95 percent speedup can be achieved by visualizing graph communities, and the visualization appears to be meaningful and reliable. The biggest graph that we examined contains above 3 million nodes and 34 million edges, and the layout computation took about five minutes. We also observed that the BigGraphVis coloring strategy can be successfully applied to produce a more informative ForceAtlas2 layout.
cs.*
In this paper, we show a new tagging scheme for cryptographic protocol messages. Under this tagging, equational theories of operators such as exclusive-or, binary addition etc. are effectively disabled, when terms are unified. We believe that this result has a significant impact on protocol analysis and security, since unification is at the heart of symbolic protocol analysis. Hence, disabling equational theories in unification implies disabling them altogether in protocol analysis for most operators and theories.
cs.*
Peer-review is a necessary and essential quality control step for scientific publications but lacks proper incentives. Indeed, the process, which is very costly in terms of time and intellectual investment, not only is not remunerated by the journals but is also not openly recognized by the academic community as a relevant scientific output for a researcher. Therefore, scientific dissemination is affected in timeliness, quality, and fairness. Here, to solve this issue, we propose a blockchain-based incentive system that rewards scientists for peer-reviewing other scientists' work and that builds up trust and reputation. We designed a privacy-oriented protocol of smart contracts called Ants-Review that allows authors to issue a bounty for open anonymous peer-reviews on Ethereum. If requirements are met, peer-reviews will be accepted and paid by the approver proportionally to their assessed quality. To promote ethical behavior and inclusiveness the system implements a gamified mechanism that allows the whole community to evaluate the peer-reviews and vote for the best ones.
cs.*
Scientific programmers often turn to vendor-tuned Basic Linear Algebra Subprograms (BLAS) to obtain portable high performance. However, many numerical algorithms require several BLAS calls in sequence, and those successive calls result in suboptimal performance. The entire sequence needs to be optimized in concert. Instead of vendor-tuned BLAS, a programmer could start with source code in Fortran or C (e.g., based on the Netlib BLAS) and use a state-of-the-art optimizing compiler. However, our experiments show that optimizing compilers often attain only one-quarter the performance of hand-optimized code. In this paper we present a domain-specific compiler for matrix algebra, the Build to Order BLAS (BTO), that reliably achieves high performance using a scalable search algorithm for choosing the best combination of loop fusion, array contraction, and multithreading for data parallelism. The BTO compiler generates code that is between 16% slower and 39% faster than hand-optimized code.
cs.*
Julia is a new language for writing data analysis programs that are easy to implement and run at high performance. Similarly, the Dynamic Distributed Dimensional Data Model (D4M) aims to clarify data analysis operations while retaining strong performance. D4M accomplishes these goals through a composable, unified data model on associative arrays. In this work, we present an implementation of D4M in Julia and describe how it enables and facilitates data analysis. Several experiments showcase scalable performance in our new Julia version as compared to the original Matlab implementation.
cs.*
LeoPARD supports the implementation of knowledge representation and reasoning tools for higher-order logic(s). It combines a sophisticated data structure layer (polymorphically typed {\lambda}-calculus with nameless spine notation, explicit substitutions, and perfect term sharing) with an ambitious multi-agent blackboard architecture (supporting prover parallelism at the term, clause, and search level). Further features of LeoPARD include a parser for all TPTP dialects, a command line interpreter, and generic means for the integration of external reasoners.
cs.*
The importance of swarm robotics systems in both academic research and real-world applications is steadily increasing. However, to reach widespread adoption, new models that ensure the secure cooperation of large groups of robots need to be developed. This work introduces a novel method to encapsulate cooperative robotic missions in an authenticated data structure known as Merkle tree. With this method, operators can provide the "blueprint" of the swarm's mission without disclosing its raw data. In other words, data verification can be separated from data itself. We propose a system where robots in a swarm, to cooperate towards mission completion, have to "prove" their integrity to their peers by exchanging cryptographic proofs. We show the implications of this approach for two different swarm robotics missions: foraging and maze formation. In both missions, swarm robots were able to cooperate and carry out sequential operations without having explicit knowledge about the mission's high-level objectives. The results presented in this work demonstrate the feasibility of using Merkle trees as a cooperation mechanism for swarm robotics systems in both simulation and real-robot experiments, which has implications for future decentralized robotics applications where security plays a crucial role such as environmental monitoring, infrastructure surveillance, and disaster management.
cs.*
Take a multiplicative monoid of sequences in which the multiplication is given by Hadamard product. The set of linear combinations of interleaving monoid elements then yields a ring. We consider such a construction for the monoid of hypergeometric sequences, yielding what we call the ring of hypergeometric-type sequences -- a subring of the ring of holonomic sequences. We present two algorithms in this setting: one for computing holonomic recurrence equations from hypergeometric-type normal forms and the other for finding products of hypergeometric-type terms. These are newly implemented commands in our Maple package $\texttt{HyperTypeSeq}$, which we also describe.
cs.*
Scientific discovery is increasingly dependent on a scientist's ability to acquire, curate, integrate, analyze, and share large and diverse collections of data. While the details vary from domain to domain, these data often consist of diverse digital assets (e.g. image files, sequence data, or simulation outputs) that are organized with complex relationships and context which may evolve over the course of an investigation. In addition, discovery is often collaborative, such that sharing of the data and its organizational context is highly desirable. Common systems for managing file or asset metadata hide their inherent relational structures, while traditional relational database systems do not extend to the distributed collaborative environment often seen in scientific investigations. To address these issues, we introduce ERMrest, a collaborative data management service which allows general entity-relationship modeling of metadata manipulated by RESTful access methods. We present the design criteria, architecture, and service implementation, as well as describe an ecosystem of tools and services that we have created to integrate metadata into an end-to-end scientific data life cycle. ERMrest has been deployed to hundreds of users across multiple scientific research communities and projects. We present two representative use cases: an international consortium and an early-phase, multidisciplinary research project.
cs.*
The Apache Accumulo database is an open source relaxed consistency database that is widely used for government applications. Accumulo is designed to deliver high performance on unstructured data such as graphs of network data. This paper tests the performance of Accumulo using data from the Graph500 benchmark. The Dynamic Distributed Dimensional Data Model (D4M) software is used to implement the benchmark on a 216-node cluster running the MIT SuperCloud software stack. A peak performance of over 100,000,000 database inserts per second was achieved which is 100x larger than the highest previously published value for any other database. The performance scales linearly with the number of ingest clients, number of database servers, and data size. The performance was achieved by adapting several supercomputing techniques to this application: distributed arrays, domain decomposition, adaptive load balancing, and single-program-multiple-data programming.
cs.*
While the public claim concern for their privacy, they frequently appear to overlook it. This disparity between concern and behaviour is known as the Privacy Paradox. Such issues are particularly prevalent on wearable devices. These products can store personal data, such as text messages and contact details. However, owners rarely use protective features. Educational games can be effective in encouraging changes in behaviour. Therefore, we developed the first privacy game for (Android) Wear OS watches. 10 participants used smartwatches for two months, allowing their high-level settings to be monitored. Five individuals were randomly assigned to our treatment group, and they played a dynamically-customised privacy-themed game. To minimise confounding variables, the other five received the same app but lacking the privacy topic. The treatment group improved their protection, with their usage of screen locks significantly increasing (p = 0.043). In contrast, 80% of the control group continued to never restrict their settings. After the posttest phase, we evaluated behavioural rationale through semi-structured interviews. Privacy concerns became more nuanced in the treatment group, with opinions aligning with behaviour. Actions appeared influenced primarily by three factors: convenience, privacy salience and data sensitivity. This is the first smartwatch game to encourage privacy-protective behaviour.
cs.*
This paper shows a cut along a crease on an origami sheet makes simple modeling of popular traditional basic folds such as a squash fold in computational origami. The cut operation can be applied to other classical folds and significantly simplify their modeling and subsequent implementation in the context of computational origami.
cs.*
In this paper, we propose a novel space partitioning strategy for implicit hierarchy visualization such that the new plot not only has a tidy layout similar to the treemap, but also is flexible to data changes similar to the Voronoi treemap. To achieve this, we define a new distance function and neighborhood relationship between sites so that space will be divided by axis-aligned segments. Then a sweepline+skyline based heuristic algorithm is proposed to allocate the partitioned spaces to form an orthogonal Voronoi diagram with orthogonal rectangles. To the best of our knowledge, it is the first time to use a sweepline-based strategy for the Voronoi treemap. Moreover, we design a novel strategy to initialize the diagram status and modify the status update procedure so that the generation of our plot is more effective and efficient. We show that the proposed algorithm has an O(nlog(n)) complexity which is the same as the state-of-the-art Voronoi treemap. To this end, we show via experiments on the artificial dataset and real-world dataset the performance of our algorithm in terms of computation time, converge rate, and aspect ratio. Finally, we discuss the pros and cons of our method and make a conclusion.
cs.*
We give an algorithm to enumerate the results on trees of monadic second-order (MSO) queries represented by nondeterministic tree automata. After linear time preprocessing (in the input tree), we can enumerate answers with linear delay (in each answer). We allow updates on the tree to take place at any time, and we can then restart the enumeration after logarithmic time in the tree. Further, all our combined complexities are polynomial in the automaton. Our result follows our previous circuit-based enumeration algorithms based on deterministic tree automata, and is also inspired by our earlier result on words and nondeterministic sequential extended variable-set automata in the context of document spanners. We extend these results and combine them with a recent tree balancing scheme by Niewerth, so that our enumeration structure supports updates to the underlying tree in logarithmic time (with leaf insertions, leaf deletions, and node relabelings). Our result implies that, for MSO queries with free first-order variables, we can enumerate the results with linear preprocessing and constant-delay and update the underlying tree in logarithmic time, which improves on several known results for words and trees. Building on lower bounds from data structure research, we also show unconditionally that up to a doubly logarithmic factor the update time of our algorithm is optimal. Thus, unlike other settings, there can be no algorithm with constant update time.
cs.*
Timing and power consumption play an important role in the design of embedded systems. Furthermore, both properties are directly related to the safety requirements of many embedded systems. With regard to availability requirements, power considerations are of uttermost importance for battery operated systems. Validation of timing and power requires observability of these properties. In many cases this is difficult, because the observability is either not possible or requires big extra effort in the system validation process. In this paper, we present a measurement-based approach for the joint timing and power analysis of Synchronous Dataflow (SDF) applications running on a shared memory multiprocessor systems-on-chip (MPSoC) architecture. As a proof-of-concept, we implement an MPSoC system with configurable power and timing measurement interfaces inside a Field Programmable Gate Array (FPGA). Our experiments demonstrate the viability of our approach being able of accurately analyzing different mappings of image processing applications (Sobel filter and JPEG encoder) on an FPGA-based MPSoC implementation.
cs.*
This paper introduces NVCache, an approach that uses a non-volatile main memory (NVMM) as a write cache to improve the write performance of legacy applications. We compare NVCache against file systems tailored for NVMM (Ext4-DAX and NOVA) and with I/O-heavy applications (SQLite, RocksDB). Our evaluation shows that NVCache reaches the performance level of the existing state-of-the-art systems for NVMM, but without their limitations: NVCache does not limit the size of the stored data to the size of the NVMM, and works transparently with unmodified legacy applications, providing additional persistence guarantees even when their source code is not available.
cs.*
Read mapping is a fundamental, yet computationally-expensive step in many genomics applications. It is used to identify potential matches and differences between fragments (called reads) of a sequenced genome and an already known genome (called a reference genome). To address the computational challenges in genome analysis, many prior works propose various approaches such as filters that select the reads that must undergo expensive computation, efficient heuristics, and hardware acceleration. While effective at reducing the computation overhead, all such approaches still require the costly movement of a large amount of data from storage to the rest of the system, which can significantly lower the end-to-end performance of read mapping in conventional and emerging genomics systems. We propose GenStore, the first in-storage processing system designed for genome sequence analysis that greatly reduces both data movement and computational overheads of genome sequence analysis by exploiting low-cost and accurate in-storage filters. GenStore leverages hardware/software co-design to address the challenges of in-storage processing, supporting reads with 1) different read lengths and error rates, and 2) different degrees of genetic variation. Through rigorous analysis of read mapping processes, we meticulously design low-cost hardware accelerators and data/computation flows inside a NAND flash-based SSD. Our evaluation using a wide range of real genomic datasets shows that GenStore, when implemented in three modern SSDs, significantly improves the read mapping performance of state-of-the-art software (hardware) baselines by 2.07-6.05$\times$ (1.52-3.32$\times$) for read sets with high similarity to the reference genome and 1.45-33.63$\times$ (2.70-19.2$\times$) for read sets with low similarity to the reference genome.
cs.*
Systems-on-Chips (SoCs) that power autonomous vehicles (AVs) must meet stringent performance and safety requirements prior to deployment. With increasing complexity in AV applications, the system needs to meet these real-time demands of multiple safety-critical applications simultaneously. A typical AV-SoC is a heterogeneous multiprocessor consisting of accelerators supported by general-purpose cores. Such heterogeneity, while needed for power-performance efficiency, complicates the art of task scheduling. In this paper, we demonstrate that hardware heterogeneity impacts the scheduler's effectiveness and that optimizing for only the real-time aspect of applications is not sufficient in AVs. Therefore, a more holistic approach is required -- one that considers global Quality-of-Mission (QoM) metrics, as defined in the paper. We then propose HetSched, a multi-step scheduler that leverages dynamic runtime information about the underlying heterogeneous hardware platform, along with the applications' real-time constraints and the task traffic in the system to optimize overall mission performance. HetSched proposes two scheduling policies: MSstat and MSdyn and scheduling optimizations like task pruning, hybrid heterogeneous ranking and rank update. HetSched improves overall mission performance on average by 4.6x, 2.6x and 2.6x when compared against CPATH, ADS and 2lvl-EDF (state-of-the-art real-time schedulers built for heterogeneous systems), respectively, and achieves an average of 53.3% higher hardware utilization, while meeting 100% critical deadlines for real-world applications of autonomous vehicles. Furthermore, when used as part of an SoC design space exploration loop, in comparison to prior schedulers, HetSched reduces the number of processing elements required by an SoC to safely complete AV's missions by 35% on average while achieving 2.7x lower energy-mission time product.
cs.*
With the increasing popularity of accelerator technologies (e.g., GPUs and TPUs) and the emergence of domain-specific computing via ASICs and FPGA, the matter of heterogeneity and understanding its ramifications on the performance has become more critical than ever before. However, it is challenging to effectively educate students about the potential impacts of heterogeneity on the performance of distributed systems; and on the logic of resource allocation methods to efficiently utilize the resources. Making use of the real infrastructure for benchmarking the performance of heterogeneous machines, for different applications, with respect to different objectives, and under various workload intensities is cost- and time-prohibitive. To reinforce the quality of learning about various dimensions of heterogeneity, and to decrease the widening gap in education, we develop an open-source simulation tool, called E2C, that can help students researchers to study any type of heterogeneous (or homogeneous) computing system and measure its performance under various configurations. E2C is equipped with an intuitive graphical user interface (GUI) that enables its users to easily examine system-level solutions (scheduling, load balancing, scalability, etc.) in a controlled environment within a short time. E2C is a discrete event simulator that offers the following features: (i) simulating a heterogeneous computing system; (ii) implementing a newly developed scheduling method and plugging it into the system, (iii) measuring energy consumption and other output-related metrics; and (iv) powerful visual aspects to ease the learning curve for students. We used E2C as an assignment in the Distributed and Cloud Computing course. Our anonymous survey study indicates that students rated E2C with the score of 8.7 out of 10 for its usefulness in understanding the concepts of scheduling in heterogeneous computing.
cs.*
Cache partitioning techniques have been successfully adopted to mitigate interference among concurrently executing real-time tasks on multi-core processors. Considering that the execution time of a cache-sensitive task strongly depends on the cache available for it to use, co-optimizing cache partitioning and task allocation improves the system's schedulability. In this paper, we propose a hybrid multi-layer design space exploration technique to solve this multi-resource management problem. We explore the interplay between cache partitioning and schedulability by systematically interleaving three optimization layers, viz., (i) in the outer layer, we perform a breadth-first search combined with proactive pruning for cache partitioning; (ii) in the middle layer, we exploit a first-fit heuristic for allocating tasks to cores; and (iii) in the inner layer, we use the well-known recurrence relation for the schedulability analysis of non-preemptive fixed-priority (NP-FP) tasks in a uniprocessor setting. Although our focus is on NP-FP scheduling, we evaluate the flexibility of our framework in supporting different scheduling policies (NP-EDF, P-EDF) by plugging in appropriate analysis methods in the inner layer. Experiments show that, compared to the state-of-the-art techniques, the proposed framework can improve the real-time schedulability of NP-FP task sets by an average of 15.2% with a maximum improvement of 233.6% (when tasks are highly cache-sensitive) and a minimum of 1.6% (when cache sensitivity is low). For such task sets, we found that clustering similar-period (or mutually compatible) tasks often leads to higher schedulability (on average 7.6%) than clustering by cache sensitivity. In our evaluation, the framework also achieves good results for preemptive and dynamic-priority scheduling policies.
cs.*
A universal deterministic inhibitor Petri net with 14 places, 29 transitions and 138 arcs was constructed via simulation of Neary and Woods' weakly universal Turing machine with 2 states and 4 symbols; the total time complexity is exponential in the running time of their weak machine. To simulate the blank words of the weakly universal Turing machine, a couple of dedicated transitions insert their codes when reaching edges of the working zone. To complete a chain of a given Petri net encoding to be executed by the universal Petri net, a translation of a bi-tag system into a Turing machine was constructed. The constructed Petri net is universal in the standard sense; a weaker form of universality for Petri nets was not introduced in this work.
cs.*
Log data anomaly detection is a core component in the area of artificial intelligence for IT operations. However, the large amount of existing methods makes it hard to choose the right approach for a specific system. A better understanding of different kinds of anomalies, and which algorithms are suitable for detecting them, would support researchers and IT operators. Although a common taxonomy for anomalies already exists, it has not yet been applied specifically to log data, pointing out the characteristics and peculiarities in this domain. In this paper, we present a taxonomy for different kinds of log data anomalies and introduce a method for analyzing such anomalies in labeled datasets. We applied our taxonomy to the three common benchmark datasets Thunderbird, Spirit, and BGL, and trained five state-of-the-art unsupervised anomaly detection algorithms to evaluate their performance in detecting different kinds of anomalies. Our results show, that the most common anomaly type is also the easiest to predict. Moreover, deep learning-based approaches outperform data mining-based approaches in all anomaly types, but especially when it comes to detecting contextual anomalies.
cs.*
CRDTs are distributed data types that make eventual consistency of a distributed object possible and non ad-hoc. Specifically, state-based CRDTs ensure convergence through disseminating the en- tire state, that may be large, and merging it to other replicas; whereas operation-based CRDTs disseminate operations (i.e., small states) assuming an exactly-once reliable dissemination layer. We introduce Delta State Conflict-Free Replicated Datatypes ({\delta}-CRDT) that can achieve the best of both worlds: small messages with an incremental nature, as in operation-based CRDTs, disseminated over unreliable communication channels, as in traditional state-based CRDTs. This is achieved by defining {\delta}-mutators to return a delta-state, typically with a much smaller size than the full state, that is joined to both: local and remote states. We introduce the {\delta}-CRDT framework, and we explain it through establishing a correspondence to current state-based CRDTs. In addition, we present an anti-entropy algorithm that ensures causal consistency, and we introduce two {\delta}-CRDT specifications of well-known replicated datatypes.
cs.*
Graph processing has become an important part of various areas of computing, including machine learning, medical applications, social network analysis, computational sciences, and others. A growing amount of the associated graph processing workloads are dynamic, with millions of edges added or removed per second. Graph streaming frameworks are specifically crafted to enable the processing of such highly dynamic workloads. Recent years have seen the development of many such frameworks. However, they differ in their general architectures (with key details such as the support for the concurrent execution of graph updates and queries, or the incorporated graph data organization), the types of updates and workloads allowed, and many others. To facilitate the understanding of this growing field, we provide the first analysis and taxonomy of dynamic and streaming graph processing. We focus on identifying the fundamental system designs and on understanding their support for concurrency, and for different graph updates as well as analytics workloads. We also crystallize the meaning of different concepts associated with streaming graph processing, such as dynamic, temporal, online, and time-evolving graphs, edge-centric processing, models for the maintenance of updates, and graph databases. Moreover, we provide a bridge with the very rich landscape of graph streaming theory by giving a broad overview of recent theoretical related advances, and by discussing which graph streaming models and settings could be helpful in developing more powerful streaming frameworks and designs. We also outline graph streaming workloads and research challenges.
cs.*
Modern high load applications store data using multiple database instances. Such an architecture requires data consistency, and it is important to ensure even distribution of data among nodes. Load balancing is used to achieve these goals. Hashing is the backbone of virtually all load balancing systems. Since the introduction of classic Consistent Hashing, many algorithms have been devised for this purpose. One of the purposes of the load balancer is to ensure storage cluster scalability. It is crucial for the performance of the whole system to transfer as few data records as possible during node addition or removal. The load balancer hashing algorithm has the greatest impact on this process. In this paper we experimentally evaluate several hashing algorithms used for load balancing, conducting both simulated and real system experiments. To evaluate algorithm performance, we have developed a benchmark suite based on Unidata MDM~ -- a scalable toolkit for various Master Data Management (MDM) applications. For assessment, we have employed three criteria~ -- uniformity of the produced distribution, the number of moved records, and computation speed. Following the results of our experiments, we have created a table, in which each algorithm is given an assessment according to the abovementioned criteria.
cs.*
The performance of today's in-memory indexes is bottlenecked by the memory latency/bandwidth wall. Processing-in-memory (PIM) is an emerging approach that potentially mitigates this bottleneck, by enabling low-latency memory access whose aggregate memory bandwidth scales with the number of PIM nodes. There is an inherent tension, however, between minimizing inter-node communication and achieving load balance in PIM systems, in the presence of workload skew. This paper presents PIM-tree, an ordered index for PIM systems that achieves both low communication and high load balance, regardless of the degree of skew in the data and the queries. Our skew-resistant index is based on a novel division of labor between the multi-core host CPU and the PIM nodes, which leverages the strengths of each. We introduce push-pull search, which dynamically decides whether to push queries to a PIM-tree node (CPU -> PIM-node) or pull the node's keys back to the CPU (PIM-node -> CPU) based on workload skew. Combined with other PIM-friendly optimizations (shadow subtrees and chunked skip lists), our PIM-tree provides high-throughput, (guaranteed) low communication, and (guaranteed) high load balance, for batches of point queries, updates, and range scans. We implement the PIM-tree structure, in addition to prior proposed PIM indexes, on the latest PIM system from UPMEM, with 32 CPU cores and 2048 PIM nodes. On workloads with 500 million keys and batches of one million queries, the throughput using PIM-trees is up to 69.7x and 59.1x higher than the two best prior methods. As far as we know these are the first implementations of an ordered index on a real PIM system.
cs.*
The paper describes the new computers architecture, the main features of which has been claimed in the Russian Federation patent 2312388 and in the US patent application 11/991331. This architecture is intended to effective support of the General Purpose Parallel Computing (GPPC), the essence of which is extremely frequent switching of threads between states of activity and states of viewed in the paper the algorithmic latency. To emphasize the same impact of the architectural latency and the algorithmic latency upon GPPC, is introduced the new notion of the generalized latency and is defined its quantitative measure - the Generalized Latency Tolerance (GLT). It is shown that a well suited for GPPC implementation architecture should have high level of GLT and is described such architecture, which is called the Virtual-Threaded Machine. This architecture originates a processor virtualization in the direction of activities virtualization, which is orthogonal to the well-known direction of memory virtualization. The key elements of the architecture are 1) the distributed fine grain representation of the architectural register file, which elements are hardware swapped through levels of a microarchitectural memory, 2) the prioritized fine grain direct hardware multiprogramming, 3) the access controlled virtual addressing and 4) the hardware driven semaphores. The composition of these features lets to introduce new styles of operating system (OS) programming, which is free of interruptions, and of applied programming with a very rare using the OS services.
cs.*
Recent NVIDIA Graphics Processing Units (GPUs) can execute multiple kernels concurrently. On these GPUs, the thread block scheduler (TBS) uses the FIFO policy to schedule their thread blocks. We show that FIFO leaves performance to chance, resulting in significant loss of performance and fairness. To improve performance and fairness, we propose use of the preemptive Shortest Remaining Time First (SRTF) policy instead. Although SRTF requires an estimate of runtime of GPU kernels, we show that such an estimate of the runtime can be easily obtained using online profiling and exploiting a simple observation on GPU kernels' grid structure. Specifically, we propose a novel Structural Runtime Predictor. Using a simple Staircase model of GPU kernel execution, we show that the runtime of a kernel can be predicted by profiling only the first few thread blocks. We evaluate an online predictor based on this model on benchmarks from ERCBench, and find that it can estimate the actual runtime reasonably well after the execution of only a single thread block. Next, we design a thread block scheduler that is both concurrent kernel-aware and uses this predictor. We implement the SRTF policy and evaluate it on two-program workloads from ERCBench. SRTF improves STP by 1.18x and ANTT by 2.25x over FIFO. When compared to MPMax, a state-of-the-art resource allocation policy for concurrent kernels, SRTF improves STP by 1.16x and ANTT by 1.3x. To improve fairness, we also propose SRTF/Adaptive which controls resource usage of concurrently executing kernels to maximize fairness. SRTF/Adaptive improves STP by 1.12x, ANTT by 2.23x and Fairness by 2.95x compared to FIFO. Overall, our implementation of SRTF achieves system throughput to within 12.64% of Shortest Job First (SJF, an oracle optimal scheduling policy), bridging 49% of the gap between FIFO and SJF.
cs.*