Add PostgreSQL result for c6a.2xlarge (2026-09-18) - #2086
alexey-milovidov merged 6 commits into
Conversation
Existing PostgreSQL results were from 2022/2025 on c6a.4xlarge only. This adds a fresh run on c6a.2xlarge (8 vCPU / 16 GiB, gp3 220GB) using PostgreSQL 17.11 via the official APT repository, following the standard benchmark.sh flow (43 queries, 3 runs each, plus the concurrent QPS follow-up). - Load time: 1835.506s (99,997,497 rows) - Data size: 95,232,178,618 bytes - Query 10, run 3 failed (null) - recorded as-is per the result format Note: the table (68 GiB on disk) does not fit in this instance's 16 GiB RAM, so most queries are dominated by disk I/O throughput on this hardware rather than query complexity. Numbers are reported as measured.
|
Results for Logs:
|
|
The run of Logs:
|
|
Results for Logs:
|
|
Results for Logs:
|
|
Results for Logs:
|
|
Results for Logs:
|
Adds a PostgreSQL 17.11 result on c6a.2xlarge (8 vCPU / 16 GiB, gp3 220GB). Ran the usual
benchmark.shpath (43 queries × 3 runs, plus the concurrent QPS follow-up). Existing PostgreSQL results are from 2022/2025 on c6a.4xlarge only; this is a smaller machine.null); left as nullSorry the MonetDB result in #2082 is still unresolved on my end. I posted a working theory for the CI failure there and will follow up separately. Didn't want that to block this PostgreSQL submission.