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 Everyone, What Is The Best Practice To Handle Disk Space On Agents? We Constantly Run Out Of The Disk Space And It Looks Like There Is No Automatic Cache Cleaner In Clearml (Pls Correct Me If I'M Wrong).

Hi everyone, what is the best practice to handle disk space on agents? We constantly run out of the disk space and it looks like there is no automatic cache cleaner in clearml (pls correct me if I'm wrong).

  
  
Posted one year ago
Votes Newest

Answers 12


ContemplativeGoat37 basically the ClearML Cache manager code makes sure there are no more than the set number of files (100 in this case) in the cache folder, and removes old files as it creates new ones

  
  
Posted one year ago

it is in number of files

  
  
Posted one year ago

I think so

  
  
Posted one year ago

ok, so that number basically represents the number of datasets, right?

  
  
Posted one year ago

is it in gb or mb?

  
  
Posted one year ago

as far as I understand files represent datasets, right? so if I have a structure like this it will count only .json files?

  
  
Posted one year ago

ok, got it, thank you

  
  
Posted one year ago

can you please elaborate on how it works? for now it's quite unclear.

  
  
Posted one year ago

Hey
There is a cache limit that you can change by modifying the conf file.
You simply add this to clearml.conf :

storage {
cache {
default_cache_manager_size: 100
}
}
(100 is the defasult value)

Depending on what you need to achieve, there are more advanced cache management tools.

  
  
Posted one year ago

The enterprise version has a more advanced cache, I think

  
  
Posted one year ago

got it, thank you. and is it the same for the enterprise version?

  
  
Posted one year ago

They actually represent whatever you downloaded locally (for example any artifact for which you called "get_local_copy")

  
  
Posted one year ago
609 Views
12 Answers
one year ago
one year ago
Tags