Repository navigation

Performance: shard layout, call-free wrapped reads, fixed index load - #6

Merged

floatdrop merged 4 commits into

mainfrom

Nov 12, 2026

Merged

Performance: shard layout, call-free wrapped reads, fixed index load#6
floatdrop merged 4 commits into
mainfrom
perf-review

Commits on Nov 12, 2026
- Lay shards out in whole cache lines
  
  floatdrop
  and
  claude
  committed
  1a60277
- Keep reads free of calls when a cache is wrapped in a generic type
  
  floatdrop
  and
  claude
  committed
  538cd5b
- Give every hash index three slots per entry
  
  floatdrop
  and
  claude
  committed
  5f5c220
- Measure throughput again, after the three changes
  
  floatdrop
  and
  claude
  committed
  f9d0d36