Re: [PoC] Improve dead tuple storage for lazy vacuum
Masahiko Sawada <sawada.mshk@gmail.com>
Attachments
- v31-0013-Add-min-and-max-classes-for-node3-and-node125.patch (application/octet-stream) patch v31-0013
- v31-0011-Remove-the-max-memory-deduction-from-TidStore.patch (application/octet-stream) patch v31-0011
- v31-0009-Review-vacuum-integration.patch (application/octet-stream) patch v31-0009
- v31-0007-Review-radix-tree.patch (application/octet-stream) patch v31-0007
- v31-0014-Revert-building-benchmark-module-for-CI.patch (application/octet-stream) patch v31-0014
- v31-0005-Tool-for-measuring-radix-tree-and-tidstore-perfo.patch (application/octet-stream) patch v31-0005
- v31-0008-Review-TidStore.patch (application/octet-stream) patch v31-0008
- v31-0003-Add-radixtree-template.patch (application/octet-stream) patch v31-0003
- v31-0006-Use-TIDStore-for-storing-dead-tuple-TID-during-l.patch (application/octet-stream) patch v31-0006
- v31-0002-Move-some-bitmap-logic-out-of-bitmapset.c.patch (application/octet-stream) patch v31-0002
- v31-0012-Revert-the-update-for-the-minimum-value-of-maint.patch (application/octet-stream) patch v31-0012
- v31-0001-Introduce-helper-SIMD-functions-for-small-byte-a.patch (application/octet-stream) patch v31-0001
- v31-0010-Radix-tree-optionally-tracks-memory-usage-when-R.patch (application/octet-stream) patch v31-0010
- v31-0004-Add-TIDStore-to-store-sets-of-TIDs-ItemPointerDa.patch (application/octet-stream) patch v31-0004
On Fri, Mar 10, 2023 at 11:30 PM Masahiko Sawada <sawada.mshk@gmail.com> wrote: > > On Fri, Mar 10, 2023 at 3:42 PM John Naylor > <john.naylor@enterprisedb.com> wrote: > > > > On Thu, Mar 9, 2023 at 1:51 PM Masahiko Sawada <sawada.mshk@gmail.com> wrote: > > > > > I've attached the new version patches. I merged improvements and fixes > > > I did in the v29 patch. > > > > I haven't yet had a chance to look at those closely, since I've had to devote time to other commitments. I remember I wasn't particularly impressed that v29-0008 mixed my requested name-casing changes with a bunch of other random things. Separating those out would be an obvious way to make it easier for me to look at, whenever I can get back to this. I need to look at the iteration changes as well, in addition to testing memory measurement (thanks for the new results, they look encouraging). > > Okay, I'll separate them again. Attached new patch series. In addition to separate them again, I've fixed a conflict with HEAD. Regards, -- Masahiko Sawada Amazon Web Services: https://aws.amazon.com
Commits
-
radixtree: Fix SIGSEGV at update of embeddable value to non-embeddable.
- bb7f195ff788 17.0 landed
-
Get rid of anonymous struct
- bf183f168c44 17.0 landed
-
Teach radix tree to embed values at runtime
- 0fe5f64367bc 17.0 landed
-
Teach TID store to skip bitmap for small numbers of offsets
- f35bd9bf359d 17.0 landed
-
Use bump context for TID bitmaps stored by vacuum
- 8a1b31e6e596 17.0 landed
-
Fix alignment of stack variable
- 0ea51bac3802 17.0 landed
-
Use TidStore for dead tuple TIDs storage during lazy vacuum.
- 667e65aac354 17.0 landed
-
Rethink create and attach APIs of shared TidStore.
- 2d8f56dabbfd 17.0 landed
-
Fix inconsistent function prototypes with function definitions.
- a0e22ef9114b 17.0 landed
-
Fix a calculation in TidStoreCreate().
- 4edb37e322a6 17.0 landed
-
Fix potential integer handling issue in radixtree.h.
- 80d5d4937c16 17.0 landed
-
Add TIDStore, to store sets of TIDs (ItemPointerData) efficiently.
- 30e144287a72 17.0 landed
-
Fix link error for test_radixtree module on Windows
- ab6ae6260372 17.0 landed
- 9552e3ace317 17.0 landed
-
Blind attempt to fix ODR violations
- 1f1d73a8b83f 17.0 landed
-
Fix incorrect format specifier for int64
- e444ebcb85c0 17.0 landed
-
Fix redefinition of typedefs
- ac234e6377dd 17.0 landed
-
Add template for adaptive radix tree
- ee1b30f128d8 17.0 landed
-
Fix signedness error in 9f225e992 for gcc
- de7c6fe8347a 17.0 landed
-
Introduce helper SIMD functions for small byte arrays
- 9f225e992bed 17.0 landed
-
Optimize vacuuming of relations with no indexes.
- c120550edb86 17.0 cited
-
Add bound check before bsearch() for performance
- bbaf315309ed 14.0 cited
-
Allocate consecutive blocks during parallel seqscans
- 56788d2156fc 14.0 cited