Skip to content

Change default number of buckets per type to 1 #1902

Closed

Description

@lvca

Currently the default number of buckets per type is 8 (but in some configuration is automatically adjusted to the number of core - 1). Even if this increase performance with multi-thread async insertion, in all the other use case creates the following issues:

  • 8X the number of open files. OSs have limitation with the number of files a process can open
  • indexes are partitioned per bucket, so unless you configure a partitioned index strategy, by default each lookup cost is multiplied by 8 buckets (one sub-index per bucket)

This proposal is to reduce to 1 the default of buckets when you create a type, so 1 type = 1 bucket = 1 file.

Activity

  1. self-assigned this
    on Jan 9, 2025
  2. added this to the 25.1.1 milestone on Jan 9, 2025
  3. added a commit that references this issue on Jan 9, 2025
    d98cd23
  4. added a commit that references this issue on Jan 13, 2025
    bc3868c
  5. added 2 commits that reference this issue on Jun 28, 2026
    911ced2
    421c64b
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

enhancementNew feature or request

Type

No type

Projects

No projects

    Milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions