uncledecart commited on
Commit
ba0f83d
·
verified ·
1 Parent(s): 90bef92

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +105 -3
README.md CHANGED
@@ -1,3 +1,105 @@
1
- ---
2
- license: apache-2.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ ---
4
+
5
+ ## Content of this dataset
6
+
7
+
8
+ This dataset contains runs of experiments to evaluate feasibility of real-time capabilities of Dell R660XS(SN-CB5GXC4) server.
9
+ There are 3 types of tests: caterpillar, cyclictest, 48-hour benchmark (for detailed explanation see below).
10
+
11
+ It has following tests runs
12
+
13
+ | folder | test | description |
14
+ |----------|-------------|------------------------|
15
+ | 05-19-35 | caterpillar | before BIOS correction |
16
+ | 06-38-03 | cyclictest | after BIOS correction |
17
+ | 15-12-05 | 48-hour | with msr error |
18
+ | 10-49-18 | 48-hour | wihout msr error |
19
+ | 11-04-24 | 48-hour | using rdtset |
20
+
21
+ For more info about analysis done check out [this](https://github.com/zededa/rt-tools) repo
22
+
23
+
24
+ ## What Is the Caterpillar Benchmark for RT?
25
+
26
+ The **Caterpillar RT Benchmark** is a diagnostic tool designed to:
27
+ - Evaluate **real-time responsiveness** of CPU cores under varying system conditions.
28
+ - Measure **timing jitter** (variations in scheduling and latency).
29
+ - Identify hardware or kernel-level contributors to timing instability (e.g., IRQs, temperature fluctuations).
30
+
31
+ The benchmark runs multiple real-time workloads while collecting detailed monitoring data from the system.
32
+
33
+ ---
34
+
35
+ ### Input Data Description
36
+
37
+ The benchmark generates several monitoring files in the **output directory**, each containing time-synchronized measurements from different subsystems.
38
+
39
+ | File | Description |
40
+ |------|--------------|
41
+ | `output.csv` | Primary timing dataset (sample and session jitter metrics). |
42
+ | `cpu_monitor.csv` | CPU temperature, frequency, and utilization per core. |
43
+ | `cpustat_monitor.csv` | Detailed CPU statistics (user/system/idle times, context switches). |
44
+ | `irq_monitor.csv` | IRQ activity counts per CPU core over time. |
45
+ | `softirq_monitor.csv` | Kernel-level soft interrupt counts per CPU core. |
46
+ | `meminfo_monitor.csv` | Memory pressure and allocation statistics. |
47
+ | `sysinfo.json` | System configuration and metadata (CPU topology, kernel version, etc.). |
48
+
49
+ ---
50
+
51
+ ### Structure of `output.csv`
52
+
53
+ The file `output.csv` contains per-sample jitter measurements recorded during the benchmark.
54
+ Each row represents a single sampling point with both **sample-level** and **session-level** timing information.
55
+
56
+ | Column | Description | Units |
57
+ |---------|--------------|-------|
58
+ | `timestamp` | Timestamp of the measurement | ISO 8601 datetime |
59
+ | `SampleMin` | Minimum latency within current sampling window | microseconds (µs) |
60
+ | `SampleMax` | Maximum latency within current sampling window | microseconds (µs) |
61
+ | `SmplJitter` | Difference between `SampleMax` and `SampleMin` (per-sample jitter) | microseconds (µs) |
62
+ | `SessionMin` | Minimum latency observed since session start | microseconds (µs) |
63
+ | `SessionMax` | Maximum latency observed since session start | microseconds (µs) |
64
+ | `SessionJitter` | `SessionMax - SessionMin`, cumulative jitter for the session | microseconds (µs) |
65
+ | `Sample` | Sequential sample index | integer |
66
+
67
+
68
+ ## What Is the cyclictest Benchmark for RT?
69
+
70
+ Cyclictest repeatedly and precisely measures the gap between a thread’s scheduled wake-up time and the moment it actually wakes from clock_nanosleep, providing detailed latency statistics. It identifies real-time latency sources across hardware, firmware, and the operating system.
71
+
72
+ for more details see [here](https://wiki.linuxfoundation.org/realtime/documentation/howto/tools/cyclictest/start)
73
+
74
+
75
+ ### Input Data Description
76
+
77
+ The benchmark generates several monitoring files in the **output directory**, each containing time-synchronized measurements from different subsystems.
78
+
79
+ | File | Description |
80
+ |------|--------------|
81
+ | `output.csv` | csv file containing test results |
82
+ | `cpu_monitor.csv` | CPU temperature, frequency, and utilization per core. |
83
+ | `cpustat_monitor.csv` | Detailed CPU statistics (user/system/idle times, context switches). |
84
+ | `irq_monitor.csv` | IRQ activity counts per CPU core over time. |
85
+ | `softirq_monitor.csv` | Kernel-level soft interrupt counts per CPU core. |
86
+ | `meminfo_monitor.csv` | Memory pressure and allocation statistics. |
87
+ | `sysinfo.json` | System configuration and metadata (CPU topology, kernel version, etc.). |
88
+
89
+ ---
90
+
91
+ ### Structure of `output.csv`
92
+
93
+ | Short | Explanation |
94
+ |--------|----------------------------------------------------------------------|
95
+ | `timestamp` | Timestamp of the measurement |
96
+ | `T` | Thread: Thread index and thread ID |
97
+ | `P` | Priority: RT thread priority |
98
+ | `I` | Interval: Intended wake up period for the latency measuring threads |
99
+ | `C` | Count: Number of times the latency was measured (iteration count) |
100
+ | `Min` | Minimum: Minimum latency that was measured |
101
+ | `Act` | Actual: Latency measured during the latest completed iteration |
102
+ | `Avg` | Average: Average latency that was measured |
103
+ | `Max` | Maximum: Maximum latency that was measured |
104
+
105
+