Examples: query, "exact match", wildcard*, wild?ard, wild*rd
Fuzzy search: cake~ (finds cakes, bake)
Term boost: "red velvet"^4, chocolate^2
Field grouping: tags:(+work -"fun-stuff")
Escaping: Escape characters +-&|!(){}[]^"~*?:\ with \, e.g. \+
Range search: properties.timestamp:[1587729413488 TO *] (inclusive), properties.title:{A TO Z}(excluding A and Z)
Combinations: chocolate AND vanilla, chocolate OR vanilla, (chocolate OR vanilla) NOT "vanilla pudding"
Field search: properties.title:"The Title" AND text
Answered
Hi

Hi
Currently the clearml tasks submitted to a queue gets executed sequentially in a FIFO mode. Is there any way to make the agent run some of this concurrently assuming there is enough resources available in the node or it can autoscale ?

  
  
Posted one year ago
pic
More
ObedientToad56
0 × 1
Report
Votes Newest

Answers 3

@<1523701087100473344:profile|SuccessfulKoala55> , it's on a single machine.

  
  
Posted one year ago
pic
More
ObedientToad56
0 × 1
Report

Hi @<1523704207914307584:profile|ObedientToad56> , where is the agent deployed in this scenario? a single machine? a k8s cluster?

  
  
Posted one year ago
pic
More
SuccessfulKoala55
0 × 1
Report

In that case, if you do not require a GPU you can use the agent in services mode (running several tasks in parallel). Otherwise scaling solution are using the AWS autoscaler, or deploying several agents, or using a k8s cluster

  
  
Posted one year ago
pic
More
SuccessfulKoala55
0 × 1
Report
Write your answer
959 Views
3 Answers
one year ago
one year ago
Tags
mlops