-
## Issue
Mapping timestamps to LSN is slow for tenats with large SLRU counts.
The slowness comes from repeatedly querying SLRUs without using vectored get. On a tenant with 190 slru
blocks this …
-
Currently, we store SLRU pages on all shards. This is done to enable all shards to do timestamp -> LSN conversions, which require walking SLRU CLog contents.
It is redundant to do this on all shar…
-
Example:
- [test_basebackup_with_high_slru_count[release-pg16-github-actions-selfhosted-10-13-30]](https://neon-github-public-dev.s3.amazonaws.com/reports/main/10495385631/index.html#suites/c62b105f3a…
-
jcsp updated
3 months ago
-
Seems to me like instead of `extend_from_slice` (= effectively memcopying the returned page images into `self.buf`), we could make `self.buf` store the `Bytes` instead, and consume them in `self.flush…
-
## Description
Make [Shared Cache](https://ydb.tech/docs/en/concepts/glossary#shared-cache) concurrent (better lock-free).
**This improvement opens us the possibility of making DataShard reads i…
-
[Over in RelStorage](https://github.com/zodb/relstorage/issues/127) we've been having a discussion based on the observation that strict LRU isn't necessarily a good cache policy for varying workloads.…
-
As of `v0.12.4`, Moka supports only the TinyLFU cache policy. However, there are some requests and plans to support other cache policies. I am going to implement the LRU policy without refactoring the…
-
Hello @x4m,
I saw that you are working on a patch related to multiXact and SLRU In this conversation:
https://www.postgresql.org/message-id/flat/2BEC2B3F-9B61-4C1D-9FB5-5FAB0F05EF86%40yandex-tea…
-
### Database name
PostgreSQL 13.3 (Debian 13.3-1.pgdg100+1) on x86_64-pc-linux-gnu, compiled by gcc (Debian 8.3.0-6) 8.3.0, 64-bit
wal-g version 0e286c83 2021.10.15_14:51:52 Postgre…