Eric Coissac 3ba26b3dc1 feat: add sparse on-disk format for presence matrices ...

The index packing API now accepts a `sparse` parameter to generate `PersistentSparseBitMatrix` files alongside existing dense matrices. The sibling cache automatically detects this format via an `is_multi.prsb` marker file and routes queries identically to the dense variant. A new `--sparse` CLI flag exposes the option, with tests verifying end-to-end pipeline correctness and storage equivalence.

2026-08-16 21:51:02 +02:00

..

sequences

docs: add coverage reference files and flag architectural drift

2026-05-23 13:44:23 +02:00

index_architecture.md

docs: update architecture and storage specs for approximate index

2026-05-23 13:54:31 +02:00

index_architecture.refs.md

docs: add coverage reference files and flag architectural drift

2026-05-23 13:44:23 +02:00

numa_partition_runner.md

refactor(numa): replace flat runner with per-node activation channels

2026-07-03 13:03:31 +02:00

numa_worker_pools.md

feat: implement NUMA-aware worker pools for merge command

2026-06-14 23:56:52 +02:00

query.md

perf: optimize k-mer queries with sparse index and run-based aggregation

2026-07-07 18:56:41 +02:00

query.refs.md

docs: add coverage reference files and flag architectural drift

2026-05-23 13:44:23 +02:00

rebuild_filter.md

feat: add memory vectors, slice traits, and column extraction methods

2026-06-17 15:03:18 +02:00

siblings.md

feat: add sparse on-disk format for presence matrices

2026-08-16 21:51:02 +02:00