Scale test: 2.47M rows + 10K vector index benchmarked

Browse Source

Benchmarks on 128GB RAM server:
- 100K candidate filter (skills+city+status): 257ms
- 1M timesheet aggregation (revenue by client): 942ms
- 800K call log cross-reference (cold leads): 642ms
- Triple JOIN recruiter performance: 487ms
- 500K email open rate aggregation: 259ms
- COUNT all 2.47M rows: 84ms
- 10K vector search (cosine similarity): ~450ms
- Embedding throughput: 49 chunks/sec via Ollama
- RAG correctly refuses to hallucinate when no match exists

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

...

This commit is contained in:

root

2026-03-27 08:31:37 -05:00

co-authored by Claude Opus 4.6

parent 26fc98c885

commit eae51977ab

24 changed files with 369 additions and 105 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

data/datasets/call_log.parquet

BIN

Binary file not shown.

data/datasets/candidates.parquet

BIN

Binary file not shown.

data/datasets/clients.parquet

BIN

Binary file not shown.

data/datasets/email_log.parquet

BIN

Binary file not shown.

data/datasets/job_orders.parquet

BIN

Binary file not shown.

data/datasets/placements.parquet

BIN

Binary file not shown.

data/datasets/timesheets.parquet

BIN

Binary file not shown.