-
Notifications
You must be signed in to change notification settings - Fork 852
experimental: try a container growth factor of 1.5x #14811
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
experimental: try a container growth factor of 1.5x #14811
Conversation
instead of 2x. we will have to see if this has an impact in our regular performance tests.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I took the time to quickly search for this. folly does this for its fbvector implementation with 1.5 as well, except for smaller capacities, where 2 is used. And what "smaller capacity" means may depend on jemalloc.
This is to exactly match the block sizes of the allocator if possible for small sizes, and start to try and re-use memory for larger sizes.
https://github.com/facebook/folly/blob/master/folly/FBVector.h#L1151
https://github.com/facebook/folly/blob/master/folly/docs/FBVector.md
…erimental-container-growth-factor-1-dot-5
…erimental-container-growth-factor-1-dot-5
Co-authored-by: Tobias Gödderz <tobias@arangodb.com>
…erimental-container-growth-factor-1-dot-5
Merging now after 3.9 branch-off... |
Scope & Purpose
Should be merged for 3.10.
As suggest by @goedderz in #14803 (comment):
Experimental: try a container growth factor of 1.5x instead of 2x. We will have to see if this has an impact in our regular performance tests.
If this does not have any measurable or a negative effect on performance, we can simply revert the changes in this PR.
Backports:
Testing & Verification