Skip to content

calculating file hashes is far too slow. #16

Open

Description

@nickzoic

Also, if this is to be run, it should only be run once, when the files are loaded, not every time the config file is saved ...

Activity

  1. nickzoic commented on Jan 15, 2026

    @nickzoic
    CollaboratorAuthor

    this continues to be a huge problem for large .fastq files.

  2. afrubin commented on Mar 12, 2026

    @afrubin

    Can we add a quick file size check for the config file saving instead to speed things up? That seems better than removing all consistency checks, but clearly calculating the hashes for all things is not working.

  3. added this to the Preprint milestone on Mar 12, 2026
  4. nickzoic commented on Mar 13, 2026

    @nickzoic
    CollaboratorAuthor

    It's hard to quantify how much effect this actually has because of caches etc. On a pretty damn fast machine running many sha256s in parallel can rip through 6.6GB of .fastq.gz in 2 seconds (20 CPU-seconds) BUT only if the cache is hot already.

    Like a lot of CountESS things, if it could do the hash check in the background it would be way less annoying.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    optimisationPerformance optimisation

    Type

    No type

    Projects

    No projects

      Milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions