Dataset Viewer
Auto-converted to Parquet Duplicate
uid
stringclasses
10 values
dataset_id
stringclasses
1 value
year
int64
2.02k
2.02k
network_scope
stringclasses
4 values
traffic_load_pct
int64
55
97
queue_depth_packets
int64
80
1.5k
active_queue_management_enabled
stringclasses
2 values
packet_drop_pct
float64
0.2
5.2
latency_ms
int64
10
130
throughput_stable
stringclasses
2 values
flow_starvation_detected
stringclasses
2 values
ops_summary
stringclasses
10 values
question
stringclasses
1 value
ground_truth_label
stringclasses
2 values
ground_truth_rationale
stringclasses
10 values
outcome_signal
stringclasses
3 values
source_citation
stringclasses
1 value
NCG-001
network-congestion-control-coherence-risk-v0.1
2,024
datacenter
65
120
yes
0.3
12
yes
no
Moderate load, AQM active, low drops and stable throughput.
Is congestion control coherent. Answer coherent or incoherent.
coherent
Control loop stable.
stable_service
Synthetic
NCG-002
network-congestion-control-coherence-risk-v0.1
2,024
datacenter
92
900
no
3.5
85
no
yes
High load, deep queues, high drops, throughput collapse.
Is congestion control coherent. Answer coherent or incoherent.
incoherent
Queue buildup and starvation.
performance_degradation
Synthetic
NCG-003
network-congestion-control-coherence-risk-v0.1
2,023
isp
70
200
yes
0.4
18
yes
no
AQM holding latency low.
Is congestion control coherent. Answer coherent or incoherent.
coherent
Flows balanced.
stable_service
Synthetic
NCG-004
network-congestion-control-coherence-risk-v0.1
2,023
isp
95
1,200
no
4.8
110
no
yes
Bufferbloat, packet drops spike, users affected.
Is congestion control coherent. Answer coherent or incoherent.
incoherent
Uncontrolled congestion.
outage_risk
Synthetic
NCG-005
network-congestion-control-coherence-risk-v0.1
2,022
enterprise
55
80
yes
0.2
10
yes
no
Low load, stable queues.
Is congestion control coherent. Answer coherent or incoherent.
coherent
Healthy network.
stable_service
Synthetic
NCG-006
network-congestion-control-coherence-risk-v0.1
2,022
enterprise
88
750
no
2.7
70
no
yes
Voice traffic starved by bulk transfer.
Is congestion control coherent. Answer coherent or incoherent.
incoherent
QoS failure.
performance_degradation
Synthetic
NCG-007
network-congestion-control-coherence-risk-v0.1
2,021
cloud
60
140
yes
0.3
15
yes
no
Cloud edge stable.
Is congestion control coherent. Answer coherent or incoherent.
coherent
Balanced load.
stable_service
Synthetic
NCG-008
network-congestion-control-coherence-risk-v0.1
2,021
cloud
97
1,500
no
5.2
130
no
yes
Traffic spike overwhelms buffers.
Is congestion control coherent. Answer coherent or incoherent.
incoherent
Collapse risk.
outage_risk
Synthetic
NCG-009
network-congestion-control-coherence-risk-v0.1
2,020
datacenter
68
160
yes
0.4
16
yes
no
Normal operation.
Is congestion control coherent. Answer coherent or incoherent.
coherent
System balanced.
stable_service
Synthetic
NCG-010
network-congestion-control-coherence-risk-v0.1
2,020
datacenter
93
1,000
no
4
95
no
yes
Packet drops and jitter spike.
Is congestion control coherent. Answer coherent or incoherent.
incoherent
Congestion unmanaged.
performance_degradation
Synthetic

What this repo is for

Detect congestion collapse before users feel it.

Covers daily operator problems:

bufferbloat

queue overflow

packet drop spikes

flow starvation

QoS failure

latency jitter

Models trained here can flag unstable traffic patterns before outages occur.

Downloads last month
8