azure / AI-benchmarking-guide
File Change Frequency

File change frequency (churn) shows the distribution of file updates (days with at least one commit).

Overview
File Change Frequency Overall
  • There are 20 files with 1,503 lines of code.
    • 0 files changed more than 100 times (0 lines of code)
    • 0 files changed 51-100 times (0 lines of code)
    • 0 files changed 21-50 times (0 lines of code)
    • 12 files changed 6-20 times (1,105 lines of code)
    • 8 files changed 1-5 times (398 lines of code)
0% | 0% | 0% | 73% | 26%
Legend:
101+
51-100
21-50
6-20
1-5

explore: grouped by folders | grouped by update frequency | data
Contributors Count Frequency Overall
  • There are 20 files with 1,503 lines of code.
    • 0 files changed by more than 25 contributors (0 lines of code)
    • 0 files changed by 11-25 contributors (0 lines of code)
    • 0 files changed by 6-10 contributors (0 lines of code)
    • 18 files changed by 2-5 contributors (1,482 lines of code)
    • 2 files changed by 1 contributor (21 lines of code)
0% | 0% | 0% | 98% | 1%
Legend:
26+
11-25
6-10
2-5
1

explore: grouped by folders | grouped by contributors count | data
File Change Frequency per File Extension
py, md, txt, sh, cfg, gitignore, json, gitattributes
File Change Frequency per Extension
The number of recorded file updates
101+
51-100
21-50
6-20
1-5
py0% | 0% | 0% | 73% | 26%
cfg0% | 0% | 0% | 0% | 100%
File Change Frequency per Logical Decomposition
primary
primary (file change frequency)
The number of recorded file updates
101+
51-100
21-50
6-20
1-5
Benchmarks0% | 0% | 0% | 69% | 30%
ROOT0% | 0% | 0% | 100% | 0%
Infra0% | 0% | 0% | 0% | 100%
Most Frequently Changed Files (Top 20)

See data for all files...

File# lines# unitscreatedlast modified# changes
(days)		# contributorsfirst
contributor		latest
contributor
NCCLBandwidth.py
in Benchmarks/NVIDIA
 84 7 2024-12-13 2025-04-24 8 4 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
LLMBenchmark.py
in Benchmarks/AMD
 93 5 2024-12-13 2025-03-04 8 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
TransferBench.py
in Benchmarks/AMD
 53 7 2024-12-13 2025-03-04 7 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
GEMMHipblasLt.py
in Benchmarks/AMD
 101 7 2025-01-27 2025-03-04 7 2 156354973+gitaumark@users.n... 156354973+gitaumark@users.n...
FlashAttention.py
in Benchmarks/AMD
 49 3 2024-12-13 2025-03-04 6 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
NVBandwidth.py
in Benchmarks/NVIDIA
 54 6 2024-12-13 2025-03-14 6 4 azureuser@ar-h200-compute.f... aiceadmin@ha-gb200-vm.sk2zp...
HBMBandwidth.py
in Benchmarks/AMD
 94 8 2024-12-13 2025-03-04 6 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
96 8 2024-12-13 2025-03-04 6 4 markgitau@microsoft.com 156354973+gitaumark@users.n...
LLMBenchmark.py
in Benchmarks/NVIDIA
 103 6 2024-12-13 2025-03-26 6 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
RCCLBandwidth.py
in Benchmarks/AMD
 105 8 2024-12-13 2025-03-04 6 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
GEMMCublasLt.py
in Benchmarks/NVIDIA
 135 6 2024-12-13 2025-04-23 6 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
138 10 2024-12-13 2025-04-21 6 5 markgitau@microsoft.com 156354973+gitaumark@users.n...
FlashAttention.py
in Benchmarks/NVIDIA
 26 2 2024-12-13 2025-02-28 5 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
FIO.py
in Benchmarks/NVIDIA
 38 2 2024-12-13 2025-02-28 4 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
FIO.py
in Benchmarks/AMD
 40 2 2024-12-13 2025-03-04 4 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
HBMBandwidth.py
in Benchmarks/NVIDIA
 123 8 2024-12-13 2025-02-28 4 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
tools.py
in Infra
 19 3 2024-09-10 2025-03-04 3 1 156354973+gitaumark@users.n... 156354973+gitaumark@users.n...
Multichase.py
in Benchmarks/NVIDIA
 36 3 2025-03-14 2025-04-21 2 2 156354973+gitaumark@users.n... 156354973+gitaumark@users.n...
cfg
transferbench.cfg
in Benchmarks/AMD
 2 - 2025-01-28 2025-01-28 1 1 156354973+gitaumark@users.n... 156354973+gitaumark@users.n...
CPUStream.py
in Benchmarks/NVIDIA
 114 8 2025-03-14 2025-03-14 1 2 156354973+gitaumark@users.n... aiceadmin@ha-gb200-vm.sk2zp...
Files With Most Contributors (Top 20)
Based on the number of unique email addresses found in commits.

See data for all files...

File# lines# unitscreatedlast modified# changes
(days)		# contributorsfirst
contributor		latest
contributor
138 10 2024-12-13 2025-04-21 6 5 markgitau@microsoft.com 156354973+gitaumark@users.n...
NCCLBandwidth.py
in Benchmarks/NVIDIA
 84 7 2024-12-13 2025-04-24 8 4 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
96 8 2024-12-13 2025-03-04 6 4 markgitau@microsoft.com 156354973+gitaumark@users.n...
NVBandwidth.py
in Benchmarks/NVIDIA
 54 6 2024-12-13 2025-03-14 6 4 azureuser@ar-h200-compute.f... aiceadmin@ha-gb200-vm.sk2zp...
LLMBenchmark.py
in Benchmarks/AMD
 93 5 2024-12-13 2025-03-04 8 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
TransferBench.py
in Benchmarks/AMD
 53 7 2024-12-13 2025-03-04 7 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
LLMBenchmark.py
in Benchmarks/NVIDIA
 103 6 2024-12-13 2025-03-26 6 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
GEMMCublasLt.py
in Benchmarks/NVIDIA
 135 6 2024-12-13 2025-04-23 6 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
RCCLBandwidth.py
in Benchmarks/AMD
 105 8 2024-12-13 2025-03-04 6 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
HBMBandwidth.py
in Benchmarks/AMD
 94 8 2024-12-13 2025-03-04 6 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
FlashAttention.py
in Benchmarks/AMD
 49 3 2024-12-13 2025-03-04 6 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
FlashAttention.py
in Benchmarks/NVIDIA
 26 2 2024-12-13 2025-02-28 5 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
HBMBandwidth.py
in Benchmarks/NVIDIA
 123 8 2024-12-13 2025-02-28 4 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
FIO.py
in Benchmarks/NVIDIA
 38 2 2024-12-13 2025-02-28 4 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
FIO.py
in Benchmarks/AMD
 40 2 2024-12-13 2025-03-04 4 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
GEMMHipblasLt.py
in Benchmarks/AMD
 101 7 2025-01-27 2025-03-04 7 2 156354973+gitaumark@users.n... 156354973+gitaumark@users.n...
Multichase.py
in Benchmarks/NVIDIA
 36 3 2025-03-14 2025-04-21 2 2 156354973+gitaumark@users.n... 156354973+gitaumark@users.n...
CPUStream.py
in Benchmarks/NVIDIA
 114 8 2025-03-14 2025-03-14 1 2 156354973+gitaumark@users.n... aiceadmin@ha-gb200-vm.sk2zp...
tools.py
in Infra
 19 3 2024-09-10 2025-03-04 3 1 156354973+gitaumark@users.n... 156354973+gitaumark@users.n...
cfg
transferbench.cfg
in Benchmarks/AMD
 2 - 2025-01-28 2025-01-28 1 1 156354973+gitaumark@users.n... 156354973+gitaumark@users.n...
Files With Least Contributors (Top 20)
Based on the number of unique email addresses found in commits.

See data for all files...

File# lines# unitscreatedlast modified# changes
(days)		# contributorsfirst
contributor		latest
contributor
tools.py
in Infra
 19 3 2024-09-10 2025-03-04 3 1 156354973+gitaumark@users.n... 156354973+gitaumark@users.n...
cfg
transferbench.cfg
in Benchmarks/AMD
 2 - 2025-01-28 2025-01-28 1 1 156354973+gitaumark@users.n... 156354973+gitaumark@users.n...
CPUStream.py
in Benchmarks/NVIDIA
 114 8 2025-03-14 2025-03-14 1 2 156354973+gitaumark@users.n... aiceadmin@ha-gb200-vm.sk2zp...
GEMMHipblasLt.py
in Benchmarks/AMD
 101 7 2025-01-27 2025-03-04 7 2 156354973+gitaumark@users.n... 156354973+gitaumark@users.n...
Multichase.py
in Benchmarks/NVIDIA
 36 3 2025-03-14 2025-04-21 2 2 156354973+gitaumark@users.n... 156354973+gitaumark@users.n...
GEMMCublasLt.py
in Benchmarks/NVIDIA
 135 6 2024-12-13 2025-04-23 6 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
HBMBandwidth.py
in Benchmarks/NVIDIA
 123 8 2024-12-13 2025-02-28 4 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
RCCLBandwidth.py
in Benchmarks/AMD
 105 8 2024-12-13 2025-03-04 6 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
LLMBenchmark.py
in Benchmarks/NVIDIA
 103 6 2024-12-13 2025-03-26 6 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
HBMBandwidth.py
in Benchmarks/AMD
 94 8 2024-12-13 2025-03-04 6 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
LLMBenchmark.py
in Benchmarks/AMD
 93 5 2024-12-13 2025-03-04 8 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
TransferBench.py
in Benchmarks/AMD
 53 7 2024-12-13 2025-03-04 7 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
FlashAttention.py
in Benchmarks/AMD
 49 3 2024-12-13 2025-03-04 6 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
FIO.py
in Benchmarks/AMD
 40 2 2024-12-13 2025-03-04 4 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
FIO.py
in Benchmarks/NVIDIA
 38 2 2024-12-13 2025-02-28 4 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
FlashAttention.py
in Benchmarks/NVIDIA
 26 2 2024-12-13 2025-02-28 5 3 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
96 8 2024-12-13 2025-03-04 6 4 markgitau@microsoft.com 156354973+gitaumark@users.n...
NCCLBandwidth.py
in Benchmarks/NVIDIA
 84 7 2024-12-13 2025-04-24 8 4 azureuser@ar-h200-compute.f... 156354973+gitaumark@users.n...
NVBandwidth.py
in Benchmarks/NVIDIA
 54 6 2024-12-13 2025-03-14 6 4 azureuser@ar-h200-compute.f... aiceadmin@ha-gb200-vm.sk2zp...
138 10 2024-12-13 2025-04-21 6 5 markgitau@microsoft.com 156354973+gitaumark@users.n...
Correlations

File Size vs. Number of Changes: 20 points

Benchmarks/NVIDIA/NCCLBandwidth.py x: 84 lines of code y: 8 # changes Benchmarks/NVIDIA/GEMMCublasLt.py x: 135 lines of code y: 6 # changes Benchmarks/NVIDIA/Multichase.py x: 36 lines of code y: 2 # changes NVIDIA_runner.py x: 138 lines of code y: 6 # changes Benchmarks/NVIDIA/LLMBenchmark.py x: 103 lines of code y: 6 # changes Benchmarks/NVIDIA/CPUStream.py x: 114 lines of code y: 1 # changes Benchmarks/NVIDIA/NVBandwidth.py x: 54 lines of code y: 6 # changes AMD_runner.py x: 96 lines of code y: 6 # changes Benchmarks/AMD/RCCLBandwidth.py x: 105 lines of code y: 6 # changes Benchmarks/AMD/HBMBandwidth.py x: 94 lines of code y: 6 # changes Benchmarks/AMD/LLMBenchmark.py x: 93 lines of code y: 8 # changes Benchmarks/AMD/GEMMHipblasLt.py x: 101 lines of code y: 7 # changes Benchmarks/AMD/FlashAttention.py x: 49 lines of code y: 6 # changes Benchmarks/AMD/FIO.py x: 40 lines of code y: 4 # changes Infra/tools.py x: 19 lines of code y: 3 # changes Benchmarks/AMD/TransferBench.py x: 53 lines of code y: 7 # changes Benchmarks/NVIDIA/FIO.py x: 38 lines of code y: 4 # changes Benchmarks/NVIDIA/FlashAttention.py x: 26 lines of code y: 5 # changes Benchmarks/NVIDIA/HBMBandwidth.py x: 123 lines of code y: 4 # changes Benchmarks/AMD/transferbench.cfg x: 2 lines of code y: 1 # changes
8.0
# changes
  min: 1.0
  average: 5.1
  25th percentile: 4.0
  median: 6.0
  75th percentile: 6.0
  max: 8.0
0 138.0
lines of code
min: 2.0 | average: 75.15 | 25th percentile: 38.5 | median: 88.5 | 75th percentile: 104.5 | max: 138.0

Number of Contributors vs. Number of Changes: 20 points

Benchmarks/NVIDIA/NCCLBandwidth.py x: 4 # contributors y: 8 # changes Benchmarks/NVIDIA/GEMMCublasLt.py x: 3 # contributors y: 6 # changes Benchmarks/NVIDIA/Multichase.py x: 2 # contributors y: 2 # changes NVIDIA_runner.py x: 5 # contributors y: 6 # changes Benchmarks/NVIDIA/CPUStream.py x: 2 # contributors y: 1 # changes Benchmarks/NVIDIA/NVBandwidth.py x: 4 # contributors y: 6 # changes Benchmarks/AMD/LLMBenchmark.py x: 3 # contributors y: 8 # changes Benchmarks/AMD/GEMMHipblasLt.py x: 2 # contributors y: 7 # changes Benchmarks/AMD/FIO.py x: 3 # contributors y: 4 # changes Infra/tools.py x: 1 # contributors y: 3 # changes Benchmarks/AMD/TransferBench.py x: 3 # contributors y: 7 # changes Benchmarks/NVIDIA/FlashAttention.py x: 3 # contributors y: 5 # changes Benchmarks/AMD/transferbench.cfg x: 1 # contributors y: 1 # changes
8.0
# changes
  min: 1.0
  average: 5.1
  25th percentile: 4.0
  median: 6.0
  75th percentile: 6.0
  max: 8.0
0 5.0
# contributors
min: 1.0 | average: 2.9 | 25th percentile: 2.25 | median: 3.0 | 75th percentile: 3.0 | max: 5.0

Number of Contributors vs. File Size: 20 points

Benchmarks/NVIDIA/NCCLBandwidth.py x: 4 # contributors y: 84 lines of code Benchmarks/NVIDIA/GEMMCublasLt.py x: 3 # contributors y: 135 lines of code Benchmarks/NVIDIA/Multichase.py x: 2 # contributors y: 36 lines of code NVIDIA_runner.py x: 5 # contributors y: 138 lines of code Benchmarks/NVIDIA/LLMBenchmark.py x: 3 # contributors y: 103 lines of code Benchmarks/NVIDIA/CPUStream.py x: 2 # contributors y: 114 lines of code Benchmarks/NVIDIA/NVBandwidth.py x: 4 # contributors y: 54 lines of code AMD_runner.py x: 4 # contributors y: 96 lines of code Benchmarks/AMD/RCCLBandwidth.py x: 3 # contributors y: 105 lines of code Benchmarks/AMD/HBMBandwidth.py x: 3 # contributors y: 94 lines of code Benchmarks/AMD/LLMBenchmark.py x: 3 # contributors y: 93 lines of code Benchmarks/AMD/GEMMHipblasLt.py x: 2 # contributors y: 101 lines of code Benchmarks/AMD/FlashAttention.py x: 3 # contributors y: 49 lines of code Benchmarks/AMD/FIO.py x: 3 # contributors y: 40 lines of code Infra/tools.py x: 1 # contributors y: 19 lines of code Benchmarks/AMD/TransferBench.py x: 3 # contributors y: 53 lines of code Benchmarks/NVIDIA/FIO.py x: 3 # contributors y: 38 lines of code Benchmarks/NVIDIA/FlashAttention.py x: 3 # contributors y: 26 lines of code Benchmarks/NVIDIA/HBMBandwidth.py x: 3 # contributors y: 123 lines of code Benchmarks/AMD/transferbench.cfg x: 1 # contributors y: 2 lines of code
138.0
lines of code
  min: 2.0
  average: 75.15
  25th percentile: 38.5
  median: 88.5
  75th percentile: 104.5
  max: 138.0
0 5.0
# contributors
min: 1.0 | average: 2.9 | 25th percentile: 2.25 | median: 3.0 | 75th percentile: 3.0 | max: 5.0