[ https://issues.apache.org/jira/browse/SPARK-24615?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16546590#comment-16546590 ]
Saisai Shao commented on SPARK-24615: ------------------------------------- Yes, currently the user is responsible for asking yarn to get resources like GPU, for example like {{--num-gpus}}. Yes, I agree with you, the concerns you mentioned above is valid. But currently the design of this Jira only targets to the task level scheduling with accelerator resources already available. It assumes that accelerator resources is already got by executor and reported back to driver. Driver will schedule the tasks based on the available resources. For Spark to communicate with cluster manager to request other resources like GPU, currently it is not covered in this design doc. Xiangrui mentioned that Spark to communicate with cluster manager should also be covered in this SPIP, so I'm still under drafting. > Accelerator-aware task scheduling for Spark > ------------------------------------------- > > Key: SPARK-24615 > URL: https://issues.apache.org/jira/browse/SPARK-24615 > Project: Spark > Issue Type: Improvement > Components: Spark Core > Affects Versions: 2.4.0 > Reporter: Saisai Shao > Assignee: Saisai Shao > Priority: Major > Labels: Hydrogen, SPIP > > In the machine learning area, accelerator card (GPU, FPGA, TPU) is > predominant compared to CPUs. To make the current Spark architecture to work > with accelerator cards, Spark itself should understand the existence of > accelerators and know how to schedule task onto the executors where > accelerators are equipped. > Current Spark’s scheduler schedules tasks based on the locality of the data > plus the available of CPUs. This will introduce some problems when scheduling > tasks with accelerators required. > # CPU cores are usually more than accelerators on one node, using CPU cores > to schedule accelerator required tasks will introduce the mismatch. > # In one cluster, we always assume that CPU is equipped in each node, but > this is not true of accelerator cards. > # The existence of heterogeneous tasks (accelerator required or not) > requires scheduler to schedule tasks with a smart way. > So here propose to improve the current scheduler to support heterogeneous > tasks (accelerator requires or not). This can be part of the work of Project > hydrogen. > Details is attached in google doc. It doesn't cover all the implementation > details, just highlight the parts should be changed. > > CC [~yanboliang] [~merlintang] -- This message was sent by Atlassian JIRA (v7.6.3#76005) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org