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 Community, When I Configure A

Hi community, when I configure a ClearML client to S3 storage backend instead of the default file server, I assume that the connection to the S3 backend happens directly to the S3 endpoint and not routed through any service from the ClearML services like API server etc. My understanding is that the ClearML client utilizes Boto3 API calls underneath the hood that directly connect to the S3 endpoint using the credential chain locally.

  • Does ClearML store any dataset outside of this S3 storage (and local storage) for preview or compression purposes?
  • Are any data packets related to model or data versioning routed through the API server? (We are pretty confident it shouldn't, but just want to be sure from a security standpoint)
    We are trying to make the case for the security of the tool with the relevant stakeholders, so would appreciate if there was any sort of documentation link that cited this and we might have missed out. Sincere apologies if we did miss it, and thanks in advance for sharing the answer. 🙂
  
  
Posted 5 months ago
Votes Newest

Answers 2


Hi @<1523701070390366208:profile|CostlyOstrich36> , I took a look at the CLI & SDK documentation for the Dataset class, but it didn't look like I had an option to control the preview. Am I looking at the wrong place? Apologies if I missed something from the documentation.

  
  
Posted 5 months ago

Hi @<1523701132025663488:profile|SlimyElephant79> , to answer your questions:

  • Does ClearML store any dataset outside of this S3 storage (and local storage) for preview or compression purposes?Some preview data might be stored inside mongodb (if it's a table for example). This of course can be disabled so no data/previews are exposed to the ClearML backend.

  • Are any data packets related to model or data versioning routed through the API server? (We are pretty confident it shouldn't, but just want to be sure from a security standpoint)Nothing is ever routed via the API server. Actions are always performed by the SDK and the backend only stores references to the data i.e. links

  
  
Posted 5 months ago
241 Views
2 Answers
5 months ago
5 months ago
Tags
Similar posts