ClearML FAQ | Hi, A Couple Of Questions, If A Worker'S Gpus Are Being Used (Not By Trains Queue, But Because Of Another Program), Would The Queue Take Another Available Worker? Does Trains-Agent Consider The Gpu Current Usage When Picking A Worker? Second, I Know That

Unanswered

Hi, A Couple Of Questions, If A Worker'S Gpus Are Being Used (Not By Trains Queue, But Because Of Another Program), Would The Queue Take Another Available Worker? Does Trains-Agent Consider The Gpu Current Usage When Picking A Worker? Second, I Know That

Hi, a couple of questions,
If a worker's GPUs are being used (not by trains queue, but because of another program), would the queue take another available worker? Does trains-agent consider the GPU current usage when picking a worker?
Second, I know that you can add a task step and override its parameters in a pipeline. But if I want to change the commit-id, will that be possible to change in the override_parameter argument as well? AgitatedDove14 mentioned it's possible to change the commit-id by exporting the task, but then I remain with a task object I cannot feed as a step to the pipeline. So what do I do in that case?

  				
Posted 
	4 years ago

					More
				  		
  Report
		
					SmarmySeaurchin8
				
					0
					 × 1

Write your answer

1K Views

0 Answers

4 years ago

one year ago

Tags

Similar posts

Another Question About The Aws Auto_Scaler, I Managed To Get The Scaler Service Running, It Listens To A Queue Called "Aws" (Yes, Very Imaginative), When I Queue A Simple Task I See That It'S Being Handled But The Task Is Stuck As Pending. When Checking T

Hi There, We Have A Task That We Are Trying To Execute Remotely, If We Initiate It From One Computer The Task Executed Remotely Successfully But If We Initiate It From Another Computer The Task Failed. We Have One Remote Machine At The Worker Queue So It

Hello Community, I Am Trying To Run A Pipeline Using Pipeline Decorater. My Components I.E. Step_One, Step_Two.... I Want To Run This Components On A Queue Default And The Pipeline On Another Queue Services. The Issue Is The Step_One Seems To Abort Automa

[Distributed Training] Hi, I Have A Clearml Setup With K8Sglue That Spins Up Pods Of 4 Gpus When Picking Tasks Off The Clearml Queue. We Would Now Want To Proceed With Multi-Node Training, And Some Of The Examples We Are Trying Are Here.

Hello I Have An Issue With The Queues. I Am Running Clearml Server + Agents In Kubernetes. Because Of That There Is A Default Internal Queue Preconfigured Called "K8S_Scheduler". I Have Defined Another Queue Called "Default" Where I Enqueue The Tasks. A

Another Question, Can I Start A Clearml-Agent Inside A Docker Container (Part Of A Pod On A K8S-Cluster), And This Agent Would Then Pull Tasks From A Queue, But The Tasks Are Also Container-Based (Using A Different Container)?

The Overview Panel Would Be Extremely Well Suited For The Task Of Selecting A Number Of Projects For Comparing Them. Another Useful Feature Would Be To Allow Adding Information (E.G. Metrics Or Metadata) To The Tooltip. Would You Consider Adding This

Hi, I Have A Worker On A Machine Using Gpus 0,1 And Another Worker On The Same Machine Using Gpus 0,1,2,3,4,5. A Worker Ran A Task On Gpus 0,1 But For Some Reason The Second Worker Started Additional Task In Queue On Gpus 0,1,2,3,4,5, Which Caused Both Of