You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@airflow.apache.org by "Aizhamal Nurmamat kyzy (JIRA)" <ji...@apache.org> on 2019/05/18 02:46:01 UTC
[jira] [Updated] (AIRFLOW-3617) Kubernetes pod operator: Add
support for specifying GPU resources/limits
[ https://issues.apache.org/jira/browse/AIRFLOW-3617?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Aizhamal Nurmamat kyzy updated AIRFLOW-3617:
--------------------------------------------
Labels: kubernetes (was: )
Component/s: (was: kubernetes)
(was: contrib)
operators
Moving to operators component, and labeling with kubernetes as part of the component refactor.
> Kubernetes pod operator: Add support for specifying GPU resources/limits
> ------------------------------------------------------------------------
>
> Key: AIRFLOW-3617
> URL: https://issues.apache.org/jira/browse/AIRFLOW-3617
> Project: Apache Airflow
> Issue Type: Improvement
> Components: operators
> Affects Versions: 1.10.1
> Environment: Example: GKE cluster with nodepool for GPU nodes.
> Reporter: Bjørn Harald Fotland
> Assignee: Divyakant Yadav
> Priority: Major
> Labels: kubernetes
>
> The airflow.contrib.kubernetes.pod.Resources class is limited to specifying CPU and memory resources/limits.
> GPU resources are not covered. [https://kubernetes.io/docs/tasks/manage-gpus/scheduling-gpus/]
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)