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

Add post on PyTables direct chunking API #12

Merged
merged 16 commits into from
Aug 26, 2024

Conversation

ivilata
Copy link
Collaborator

@ivilata ivilata commented Aug 23, 2024

The development of this feature was funded by the NumFOCUS Foundation under a Small Development Grant.

@ivilata ivilata self-assigned this Aug 23, 2024
Copy link
Member

@FrancescAlted FrancescAlted left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Overall, the blog looks excellent to me! I just have some minor suggestions (that you may want to ignore if you don't agree with them). Well done!

posts/pytables-direct-chunking.rst Outdated Show resolved Hide resolved
posts/pytables-direct-chunking.rst Outdated Show resolved Hide resolved
posts/pytables-direct-chunking.rst Outdated Show resolved Hide resolved
:width: 75%
:align: center

In this case direct chunking yields 4.5x write and 1.9x read speedups, with write/read speeds of 0.8 GB/s and 1.6 GB/s. The absolute numbers are of course not as impressive, but the performance is still much better than that of the regular mechanism, especially when writing. Please note that the M1 CPU has a hybrid efficiency+performance core configuration. We also run the tests on another hybrid CPU of the high-performance range, in this case an Intel Core i9-13900K CPU with 8+16 cores and 32 MB L2 cache, clocked at 5.7 GHz:
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

To be honest, I'd not add so many benchmarks of different machines, as this can overwhelm readers a bit ;-). I'd skip the Intel machine, as it has a similar architecture (and behavior) than the AMD.

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yup, I agree, though the write results on the Intel are IMO too good not to mention them, so I replaced the whole passage with a short summary sentence in commit fa0dc00. Does it look ok to you like this?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, that's fine.

Copy link
Member

@avalentino avalentino left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM
Thanks @ivilata

@ivilata
Copy link
Collaborator Author

ivilata commented Aug 26, 2024

Thanks @FrancescAlted and @avalentino for the review!

@ivilata ivilata merged commit 1e24006 into Blosc:master Aug 26, 2024
2 checks passed
@ivilata ivilata deleted the pytables-direct-chunking branch August 26, 2024 09:19
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants