Charles KWON OhJun 96d72a456c perf: CDX zero-alloc internal node seek — SEEK 45% faster ...

Internal node traversal: read directly from mmap/buf slice
- No DecodeIntKeys allocation (was nKeys+1 IntKeyEntry structs)
- No key byte slice copy (compare directly against buf)
- Big-endian child/recNo read inline

CDX 50K benchmark:
  SEEK NAME: 362ms → 199ms (45% faster)
  SEEK ID:   320ms → 184ms (42% faster)
  SCAN:      14ms (unchanged — leaf cache handles this)
  SCOPE:     20ms → 14ms

Harbour comparison:
  SEEK: 27ms (Harbour) vs 199ms (Five) = 7.4x
  SCAN: 6ms (Harbour) vs 14ms (Five) = 2.3x

CDX cross-read: 18/18 PASS.

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

2026-04-07 14:05:20 +09:00

..

cdx

perf: CDX zero-alloc internal node seek — SEEK 45% faster

2026-04-07 14:05:20 +09:00

dbf

perf: REPLACE remove Flush + bulk build + deferred write = 1600x faster

2026-04-07 09:22:05 +09:00

mem

feat: SET DELETED filtering, SEEK/LOCATE/CONTINUE, SET command codegen

2026-04-02 22:33:59 +09:00

ntx

perf: proper 3-level bulk build — INDEX 50K: 180ms → 28ms (6.4x)

2026-04-07 12:08:27 +09:00

base.go

feat: SET DELETED filtering, SEEK/LOCATE/CONTINUE, SET command codegen

2026-04-02 22:33:59 +09:00

bitmap_test.go

Five v0.9 — Harbour + Go fusion language

2026-03-31 09:41:50 +09:00

bitmap.go

Five v0.9 — Harbour + Go fusion language

2026-03-31 09:41:50 +09:00

driver.go

feat: SET DELETED filtering, SEEK/LOCATE/CONTINUE, SET command codegen

2026-04-02 22:33:59 +09:00

workarea.go

Five v0.9 — Harbour + Go fusion language

2026-03-31 09:41:50 +09:00