github/intel-device-plugins-for-kubernetes
7
0
Fork 0
mirror of https://github.com/intel/intel-device-plugins-for-kubernetes.git synced 2025-06-03 03:59:37 +00:00
Code Issues Packages Projects Releases Wiki Activity

gpu: add notes about gpu-plugin modes

Browse Source 
Fixes: #1381

Co-authored-by: Eero Tamminen <eero.t.tamminen@intel.com>
Signed-off-by: Tuomas Katila <tuomas.katila@intel.com>
This commit is contained in:
Tuomas Katila 2023-04-18 14:07:25 +03:00
parent 4a4a0e5d2f
commit 8971280215
1 changed files with 14 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
cmd/gpu_plugin/README.md
View File

@ -4,6 +4,7 @@ Table of Contents
* [Introduction](#introduction)
* [Modes and Configuration Options](#modes-and-configuration-options)
* [Operation modes for different workload types](#operation-modes-for-different-workload-types)
* [Installation](#installation)
* [Prerequisites](#prerequisites)
* [Drivers for discrete GPUs](#drivers-for-discrete-gpus)
@ -50,11 +51,23 @@ backend libraries can offload compute operations to GPU.
| -enable-monitoring | - | disabled | Enable 'i915_monitoring' resource that provides access to all Intel GPU devices on the node |
| -resource-manager | - | disabled | Enable fractional resource management, [see also dependencies](#fractional-resources) |
| -shared-dev-num | int | 1 | Number of containers that can share the same GPU device |
| -allocation-policy | string | none | 3 possible values: balanced, packed, none. It is meaningful when shared-dev-num > 1, balanced mode is suitable for workload balance among GPU devices, packed mode is suitable for making full use of each GPU device, none mode is the default. Allocation policy does not have effect when resource manager is enabled. |
| -allocation-policy | string | none | 3 possible values: balanced, packed, none. For shared-dev-num > 1: _balanced_ mode spreads workloads among GPU devices, _packed_ mode fills one GPU fully before moving to next, and _none_ selects first available device from kubelet. Default is _none_. Allocation policy does not have an effect when resource manager is enabled. |
The plugin also accepts a number of other arguments (common to all plugins) related to logging.
Please use the -h option to see the complete list of logging related options.
## Operation modes for different workload types
Intel GPU-plugin supports a few different operation modes. Depending on the workloads the cluster is running, some modes make more sense than others. Below is a table that explains the differences between the modes and suggests workload types for each mode. Mode selection applies to the whole GPU plugin deployment, so it is a cluster wide decision.
| Mode | Sharing | Intended workloads | Suitable for time critical workloads |
|:---- |:-------- |:------- |:------- |
| shared-dev-num == 1 | No, 1 container per GPU | Workloads using all GPU capacity, e.g. AI training | Yes |
| shared-dev-num > 1 | Yes, >1 containers per GPU | (Batch) workloads using only part of GPU resources, e.g. inference, media transcode/analytics, or CPU bound GPU workloads | No |
| shared-dev-num > 1 && resource-management | Yes and no, 1>= containers per GPU | Any. For best results, all workloads should declare their expected GPU resource usage (memory, millicores). Requires [GAS](https://github.com/intel/platform-aware-scheduling/tree/master/gpu-aware-scheduling). See also [fractional use](#fractional-resources-details) | Yes. 1000 millicores = exclusive GPU usage. See note below. |
> **Note**: Exclusive GPU usage with >=1000 millicores requires that also *all other GPU containers* specify (non-zero) millicores resource usage.
## Installation
The following sections detail how to obtain, build, deploy and test the GPU device plugin.