Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Parquet cache memory management and back-pressure #25707

Open
hiltontj opened this issue Dec 24, 2024 · 2 comments
Open

Parquet cache memory management and back-pressure #25707

hiltontj opened this issue Dec 24, 2024 · 2 comments
Assignees
Labels

Comments

@hiltontj
Copy link
Contributor

Problem statement

The parquet cache is limited in its ability to prevent an influx of cache requests from causing overuse of system memory.

Proposed solution(s)

  • Require file size in the CacheRequest so that the cache oracle can decide if it needs to prune before fetching a file, or if the file is too big for the cache entirely
  • Provide an EvictionRequest to explicitly evict entries from the cache that are no longer needed
@hiltontj hiltontj added the v3 label Dec 24, 2024
@hamzzy
Copy link

hamzzy commented Dec 24, 2024

can i work on this @hiltontj

@hiltontj hiltontj self-assigned this Dec 24, 2024
@hiltontj
Copy link
Contributor Author

@hamzzy - we appreciate your offer to contribute. There are some downstream changes that depend on this work and so we are currently hashing out the details of those. That may shape the requirements for this work, so for now, I would hold off working on this. If you keep an eye on our issue tracker for issues with the good-first-issue label, those would be better candidates to pick up if you are looking to contribute. If I end up refining this or breaking it down into separate issues, they may be good as first issues.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

2 participants