test(benchmarks): vendor benchmark suite and add write-load harness - #1021

Merged

phernandez merged 19 commits into

mainfrom

benchmark-suite

Jun 25, 2026

Commits on Jun 24, 2026

chore(benchmarks): vendor basic-memory-benchmarks suite into benchmarks/

phernandez
and
claude
committed
521bd91
feat(benchmarks): add write-path load benchmark with per-ref venv harness

phernandez
and
claude
committed
69e5b7f
docs(benchmarks): record main vs PR-1002 write-load comparison

phernandez
and
claude
committed
16258dc
docs(benchmarks): coalescing fix did not close the write-load gap

phernandez
and
claude
committed
c075bca
docs(benchmarks): materialization deferral validates the async-accept hypothesis

phernandez
and
claude
committed
4ee15d3
feat(benchmarks): add postgres backend + high-concurrency scaling to write-load

phernandez
and
claude
committed
472e4e1
docs(benchmarks): high-concurrency scaling + SQLite-vs-Postgres findings

phernandez
and
claude
committed
d018eb1
docs(benchmarks): worker pool fixes the high-concurrency wall

phernandez
and
claude
committed
2c74ef5
docs(benchmarks): Postgres pool + default-project seed fix the collapse

phernandez
and
claude
committed
8cadadf
docs(benchmarks): 3x variance bounds confirm the fixes are stable

phernandez
and
claude
committed
3b117f7
docs(benchmarks): controlled 2x2 main (direct) vs branch (async)

phernandez
and
claude
committed
8e77219
docs(benchmarks): split the 2x2 into readable per-metric rows

phernandez
and
claude
committed
1ac8777
docs(benchmarks): flip 2x2 to C-rows with main/branch/speedup columns

phernandez
and
claude
committed
790b7be
docs(benchmarks): express the 2x2 improvement as percentages

phernandez
and
claude
committed
67ae6b1
feat(benchmarks): measure time_to_embedded (semantic vector-sync drain)

phernandez
and
claude
committed
657fc13
docs(benchmarks): time_to_embedded result — embedding keeps pace

phernandez
and
claude
committed
0d3eb74
docs(benchmarks): scope embedding to local fastembed

phernandez
and
claude
committed
bccda12
docs(benchmarks): SQLite PRAGMA sweep — no measurable gain

phernandez
and
claude
committed
5cdee17

Commits on Jun 25, 2026

ci: add benchmarks scope to PR title checker

phernandez
and
claude
committed
48887f5

Navigation Menu

Uh oh!

test(benchmarks): vendor benchmark suite and add write-load harness - #1021

test(benchmarks): vendor benchmark suite and add write-load harness#1021
phernandez merged 19 commits into
mainfrom
benchmark-suite

Commits on Jun 24, 2026

chore(benchmarks): vendor basic-memory-benchmarks suite into benchmarks/

feat(benchmarks): add write-path load benchmark with per-ref venv harness

docs(benchmarks): record main vs PR-1002 write-load comparison

docs(benchmarks): coalescing fix did not close the write-load gap

docs(benchmarks): materialization deferral validates the async-accept hypothesis

feat(benchmarks): add postgres backend + high-concurrency scaling to write-load

docs(benchmarks): high-concurrency scaling + SQLite-vs-Postgres findings

docs(benchmarks): worker pool fixes the high-concurrency wall

docs(benchmarks): Postgres pool + default-project seed fix the collapse

docs(benchmarks): 3x variance bounds confirm the fixes are stable

docs(benchmarks): controlled 2x2 main (direct) vs branch (async)

docs(benchmarks): split the 2x2 into readable per-metric rows

docs(benchmarks): flip 2x2 to C-rows with main/branch/speedup columns

docs(benchmarks): express the 2x2 improvement as percentages

feat(benchmarks): measure time_to_embedded (semantic vector-sync drain)

docs(benchmarks): time_to_embedded result — embedding keeps pace

docs(benchmarks): scope embedding to local fastembed

docs(benchmarks): SQLite PRAGMA sweep — no measurable gain

Commits on Jun 25, 2026

ci: add benchmarks scope to PR title checker