Allow using a custom Spark Resource Name for a GPU #4024
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: Thomas Graves <[email protected]>
tgravescs
requested review from
GaryShen2008,
jlowe,
NvTimLiu and
revans2
as code owners
tgravescs
changed the title
|
build |
jlowe
approved these changes
Nov 3, 2021
gerashegalov
reviewed
Nov 3, 2021
| val gpu = new ResourceInformation("gpu", Array("3")) | ||
| val resources = Map("gpu" -> gpu) | ||
| val gpuAddr = GpuDeviceManager.getGPUAddrFromResources(resources, conf) | ||
| assert(gpuAddr.nonEmpty) |
There was a problem hiding this comment.
nit: could use https://www.scalatest.org/user_guide/using_OptionValues
gerashegalov
approved these changes
Nov 3, 2021
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This covers the case if someone uses a custom resource name with Spark that is actually a GPU it lets the plugin discover that resource.
For instance instead of using the standard gpu scheduling config: spark.executor.resource.gpu.amount=1, instead they make a custom GPU resource named nvidia/gpu and specify that like spark.executor.resource.nvidia/gpu.amount=1.
Our device plugin specifically was expecting the spark resource to be called "gpu". This adds a config to allow it to be changed.
Specifically adding this to later be used with MIG support for Spark on YARN. relates to #4023
More documentation in the MIG context will come later.