Skip to content

Rollback driver change for MLPerf #326

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
yangpanMS merged 2 commits into from
May 23, 2024
Merged

Rollback driver change for MLPerf #326

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
60913c3
Revert "Fixing Nvidia Drivers to latest (#312)"
yangpanMS May 23, 2024
49cf61c
Revert "Fix unit test break for mlperf (#321)"
yangpanMS May 23, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion .pipelines/azure-pipelines-linux.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,7 +16,7 @@ resources:
options: --entrypoint=""

variables:
VcVersion : 1.14.33
VcVersion : 1.14.34
ROOT: $(Build.SourcesDirectory)
CDP_DEFINITION_BUILD_COUNT: $[counter('', 0)] # needed for onebranch.pipeline.version task https://aka.ms/obpipelines/versioning
ENABLE_PRS_DELAYSIGN: 1
Expand Down
2 changes: 1 addition & 1 deletion .pipelines/azure-pipelines.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,7 +18,7 @@ pool:
vmImage: windows-latest

variables:
VcVersion : 1.14.33
VcVersion : 1.14.34
ROOT: $(Build.SourcesDirectory)
CDP_DEFINITION_BUILD_COUNT: $[counter('', 0)] # needed for onebranch.pipeline.version task https://aka.ms/obpipelines/versioning
ENABLE_PRS_DELAYSIGN: 1
Expand Down
4 changes: 2 additions & 2 deletions src/VirtualClient/VirtualClient.Actions.FunctionalTests/MLPerfProfileTests.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -46,8 +46,8 @@ public async Task MLPerfWorkloadProfileExecutesTheExpectedDependenciesAndReboot(
{
$"sudo apt update",
$"sudo apt install build-essential -yq",
$"sudo wget https://developer.download.nvidia.com/compute/cuda/12.4.1/local_installers/cuda_12.4.1_550.54.15_linux.run",
$"sudo sh cuda_12.4.1_550.54.15_linux.run --silent",
$"sudo wget https://developer.download.nvidia.com/compute/cuda/12.0.0/local_installers/cuda_12.0.0_525.60.13_linux.run",
$"sudo sh cuda_12.0.0_525.60.13_linux.run --silent",
$"sudo bash -c \"echo 'export PATH=/usr/local/cuda-11.6/bin${{PATH:+:${{PATH}}}}' | sudo tee -a /home/[a-z]+/.bashrc\"",
$"bash -c \"echo 'export LD_LIBRARY_PATH=/usr/local/cuda-11.6/lib64${{LD_LIBRARY_PATH:+:${{LD_LIBRARY_PATH}}}}' " +
"| sudo tee -a /home/[a-z]+/.bashrc\""
Expand Down
4 changes: 2 additions & 2 deletions src/VirtualClient/VirtualClient.Actions.FunctionalTests/SuperBenchmarkProfileTests.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -46,8 +46,8 @@ public async Task SuperBenchmarkWorkloadProfileExecutesTheExpectedDependenciesAn
{
$"sudo apt update",
$"sudo apt install build-essential -yq",
$"sudo wget https://developer.download.nvidia.com/compute/cuda/12.4.1/local_installers/cuda_12.4.1_550.54.15_linux.run",
$"sudo sh cuda_12.4.1_550.54.15_linux.run --silent",
$"sudo wget https://developer.download.nvidia.com/compute/cuda/12.0.0/local_installers/cuda_12.0.0_525.60.13_linux.run",
$"sudo sh cuda_12.0.0_525.60.13_linux.run --silent",
$"sudo bash -c \"echo 'export PATH=/usr/local/cuda-12.0/bin${{PATH:+:${{PATH}}}}' | sudo tee -a /home/[a-z]+/.bashrc\"",
$"bash -c \"echo 'export LD_LIBRARY_PATH=/usr/local/cuda-12.0/lib64${{LD_LIBRARY_PATH:+:${{LD_LIBRARY_PATH}}}}' | " +
$"sudo tee -a /home/[a-z]+/.bashrc\""
Expand Down
2 changes: 1 addition & 1 deletion ...ualClient/VirtualClient.Dependencies.UnitTests/CUDAAndNvidiaGPUDriverInstallationTests.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -37,7 +37,7 @@ public class CUDAAndNvidiaGPUDriverInstallationTests
$"sudo tee -a /home/anyuser/.bashrc\"";

private const string UpgradeCommand = "apt upgrade -y";
private const string InstallDriverCommand = "apt install nvidia-driver-510-server nvidia-dkms-510-server -y";
private const string InstallDriverCommand = "apt install nvidia-driver-510 nvidia-dkms-510 -y";
private const string InstallFabricManagerCommand = "apt install cuda-drivers-fabricmanager-510 -y";

private MockFixture fixture;
Expand Down
2 changes: 1 addition & 1 deletion src/VirtualClient/VirtualClient.Dependencies/CUDAAndNvidiaGPUDriverInstallation.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -343,7 +343,7 @@ private List<string> VersionSpecificInstallationCommands(LinuxDistribution linux
case LinuxDistribution.Ubuntu:
commands.Add("apt update");
commands.Add("apt upgrade -y");
commands.Add($"apt install nvidia-driver-{this.LinuxDriverVersion}-server nvidia-dkms-{this.LinuxDriverVersion}-server -y");
commands.Add($"apt install nvidia-driver-{this.LinuxDriverVersion} nvidia-dkms-{this.LinuxDriverVersion} -y");
commands.Add($"apt install cuda-drivers-fabricmanager-{this.LinuxDriverVersion} -y");

break;
Expand Down
7 changes: 3 additions & 4 deletions src/VirtualClient/VirtualClient.Main/profiles/PERF-GPU-MLPERF-NVIDIA.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,16 +4,15 @@
"RecommendedMinimumExecutionTime": "10:00:00",
"SupportedPlatforms": "linux-x64",
"SupportedOperatingSystems": "Ubuntu",
"RecommendOperationSystem": "Ubunutu-22.04, Default LinusCudaVersion,LinuxDriverVersion,LinuxLocalRunFile support this OS only.",
"SupportedRequireCustomSystemSupport": "A100_PCIe_80GBx4",
"SpecialRequirements": "This is an Nvidia GPU-specialized workload. It depends upon the system having an Nvidia GPU card/chip."
},
"Parameters": {
"Username": "",
"DiskFilter": "SizeGreaterThan:1000gb",
"LinuxCudaVersion": "12.4",
"LinuxDriverVersion": "550",
"LinuxLocalRunFile": "https://developer.download.nvidia.com/compute/cuda/12.4.1/local_installers/cuda_12.4.1_550.54.15_linux.run",
"LinuxCudaVersion": "12.0",
"LinuxDriverVersion": "525",
"LinuxLocalRunFile": "https://developer.download.nvidia.com/compute/cuda/12.0.0/local_installers/cuda_12.0.0_525.60.13_linux.run",
"RequireCustomSystemSupport": true
},
"Actions": [
Expand Down
7 changes: 3 additions & 4 deletions src/VirtualClient/VirtualClient.Main/profiles/PERF-GPU-MLPERF-TRAINING-NVIDIA.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,17 +4,16 @@
"RecommendedMinimumExecutionTime": "10:00:00",
"SupportedPlatforms": "linux-x64",
"SupportedOperatingSystems": "Ubuntu",
"RecommendOperationSystem": "Ubunutu-22.04, Default LinusCudaVersion,LinuxDriverVersion,LinuxLocalRunFile support this OS only.",
"SpecialRequirements": "This is an Nvidia GPU-specialized workload. It depends upon the system having an Nvidia GPU card/chip. Requires an OSDisk of >=256GB size and Data disk of >=8TB size",
"SupportedImplementations": "Visit the link https://github.com/mlcommons/training_results_v2.1/tree/main/NVIDIA/benchmarks enter the implementation folder to see supported implementations.",
"SupportedConfigFiles": "Visit the link https://github.com/mlcommons/training_results_v2.1/tree/main/NVIDIA/benchmarks/bert/implementations/pytorch-22.09 for each benchmark and implementation you can the supported "
},
"Parameters": {
"Username": null,
"DiskFilter": "BiggestSize",
"LinuxCudaVersion": "12.4",
"LinuxDriverVersion": "550",
"LinuxLocalRunFile": "https://developer.download.nvidia.com/compute/cuda/12.4.1/local_installers/cuda_12.4.1_550.54.15_linux.run"
"LinuxCudaVersion": "12.0",
"LinuxDriverVersion": "525",
"LinuxLocalRunFile": "https://developer.download.nvidia.com/compute/cuda/12.0.0/local_installers/cuda_12.0.0_525.60.13_linux.run"
},
"Actions": [
{
Expand Down
7 changes: 3 additions & 4 deletions src/VirtualClient/VirtualClient.Main/profiles/PERF-GPU-MLPERF.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,16 +4,15 @@
"RecommendedMinimumExecutionTime": "10:00:00",
"SupportedPlatforms": "linux-x64",
"SupportedOperatingSystems": "Ubuntu",
"RecommendOperationSystem": "Ubunutu-22.04, Default drivers support this OS only.",
"SupportedRequireCustomSystemSupport": "A100_PCIe_80GBx4",
"SpecialRequirements": "This is an Nvidia GPU-specialized workload. It depends upon the system having an Nvidia GPU card/chip."
},
"Parameters": {
"Username": "",
"DiskFilter": "SizeGreaterThan:1000gb",
"LinuxCudaVersion": "12.4",
"LinuxDriverVersion": "550",
"LinuxLocalRunFile": "https://developer.download.nvidia.com/compute/cuda/12.4.1/local_installers/cuda_12.4.1_550.54.15_linux.run",
"LinuxCudaVersion": "12.0",
"LinuxDriverVersion": "525",
"LinuxLocalRunFile": "https://developer.download.nvidia.com/compute/cuda/12.0.0/local_installers/cuda_12.0.0_525.60.13_linux.run",
"RequireCustomSystemSupport": true
},
"Actions": [
Expand Down
7 changes: 3 additions & 4 deletions src/VirtualClient/VirtualClient.Main/profiles/PERF-GPU-SUPERBENCH-NVIDIA.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,15 +5,14 @@
"RecommendedMinimumExecutionTime": "08:00:00",
"SupportedPlatforms": "linux-x64",
"SupportedOperatingSystems": "Ubuntu",
"RecommendOperationSystem": "Ubunutu-22.04, Default LinusCudaVersion,LinuxDriverVersion,LinuxLocalRunFile support this OS only.",
"SpecialRequirements": "This is an Nvidia GPU-specialized workload. It depends upon the system having an Nvidia GPU card/chip."
},
"Parameters": {
"ConfigurationFile": "default.yaml",
"Username": "",
"LinuxCudaVersion": "12.4",
"LinuxDriverVersion": "550",
"LinuxLocalRunFile": "https://developer.download.nvidia.com/compute/cuda/12.4.1/local_installers/cuda_12.4.1_550.54.15_linux.run"
"LinuxCudaVersion": "12.0",
"LinuxDriverVersion": "525",
"LinuxLocalRunFile": "https://developer.download.nvidia.com/compute/cuda/12.0.0/local_installers/cuda_12.0.0_525.60.13_linux.run"
},
"Actions": [
{
Expand Down
7 changes: 3 additions & 4 deletions src/VirtualClient/VirtualClient.Main/profiles/PERF-GPU-SUPERBENCH.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,15 +5,14 @@
"RecommendedMinimumExecutionTime": "08:00:00",
"SupportedPlatforms": "linux-x64",
"SupportedOperatingSystems": "Ubuntu",
"RecommendOperationSystem": "Ubunutu-22.04, Default LinuxCudaVersion,LinuxDriverVersion,LinuxLocalRunFile support this OS only.",
"SpecialRequirements": "This is an Nvidia GPU-specialized workload. It depends upon the system having an Nvidia GPU card/chip."
},
"Parameters": {
"ConfigurationFile": "default.yaml",
"Username": "",
"LinuxCudaVersion": "12.4",
"LinuxDriverVersion": "550",
"LinuxLocalRunFile": "https://developer.download.nvidia.com/compute/cuda/12.4.1/local_installers/cuda_12.4.1_550.54.15_linux.run"
"LinuxCudaVersion": "12.0",
"LinuxDriverVersion": "525",
"LinuxLocalRunFile": "https://developer.download.nvidia.com/compute/cuda/12.0.0/local_installers/cuda_12.0.0_525.60.13_linux.run"
},
"Actions": [
{
Expand Down