liuyu/pgvector
1
0
Fork 0
mirror of https://github.com/pgvector/pgvector.git synced 2026-07-13 08:06:54 +08:00
Code Issues Packages Projects Releases Wiki Activity

Improved cost code [skip ci]

Browse Source
This commit is contained in:
Andrew Kane
2024-09-25 17:23:29 -07:00
parent 73c5145b77
commit 058248fdcc
1 changed files with 11 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files

43
src/hnsw.c
View File

@@ -102,17 +102,13 @@ hnswbuildphasename(int64 phasenum)
} }
/* /*
* Estimate extra tuples for iterative scans * Estimate ef needed for iterative scans
*/ */
static double static int
EstimateExtraTuples(PlannerInfo *root, IndexPath *path, int m, double scalingFactor) EstimateEf(PlannerInfo *root, IndexPath *path)
{ {
double selectivity = 1; double selectivity = 1;
ListCell *lc; ListCell *lc;
int ef;
int tuplesMax;
double layerSelectivity;
double tuples;
/* Cannot estimate without limit */ /* Cannot estimate without limit */
/* limit_tuples includes offset */ /* limit_tuples includes offset */
@@ -129,26 +125,7 @@ EstimateExtraTuples(PlannerInfo *root, IndexPath *path, int m, double scalingFac
selectivity *= rinfo->norm_selec; selectivity *= rinfo->norm_selec;
} }
/* Estimate the candidates needed */ return root->limit_tuples / Max(selectivity, 0.00001);
ef = root->limit_tuples / Max(selectivity, 0.00001);
/* Remove candidates from initial scan */
ef -= hnsw_ef_search;
/* Likely not needed */
if (ef <= 0)
return 0;
/* TODO DRY with hnswcostestimate */
tuplesMax = HnswGetLayerM(m, 0) * ef;
layerSelectivity = (scalingFactor * log(path->indexinfo->tuples + 1)) / (log(m) * (1 + log(ef)));
tuples = tuplesMax * layerSelectivity;
/* Limit to ef_stream */
if (hnsw_ef_stream != -1)
tuples = Min(tuples, hnsw_ef_stream);
return tuples;
} }
/* /*
@@ -162,6 +139,7 @@ hnswcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
{ {
GenericCosts costs; GenericCosts costs;
int m; int m;
int ef;
int entryLevel; int entryLevel;
int layer0TuplesMax; int layer0TuplesMax;
double layer0Selectivity; double layer0Selectivity;
@@ -186,6 +164,8 @@ hnswcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
HnswGetMetaPageInfo(index, &m, NULL); HnswGetMetaPageInfo(index, &m, NULL);
index_close(index, NoLock); index_close(index, NoLock);
ef = hnsw_streaming ? Max(hnsw_ef_search, EstimateEf(root, path)) : hnsw_ef_search;
/* /*
* HNSW cost estimation follows a formula that accounts for the total * HNSW cost estimation follows a formula that accounts for the total
* number of tuples indexed combined with the parameters that most * number of tuples indexed combined with the parameters that most
@@ -214,16 +194,15 @@ hnswcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
* "scalingFactor" (currently hardcoded). * "scalingFactor" (currently hardcoded).
*/ */
entryLevel = (int) (log(path->indexinfo->tuples + 1) * HnswGetMl(m)); entryLevel = (int) (log(path->indexinfo->tuples + 1) * HnswGetMl(m));
layer0TuplesMax = HnswGetLayerM(m, 0) * hnsw_ef_search; layer0TuplesMax = HnswGetLayerM(m, 0) * ef;
layer0Selectivity = (scalingFactor * log(path->indexinfo->tuples + 1)) / layer0Selectivity = (scalingFactor * log(path->indexinfo->tuples + 1)) /
(log(m) * (1 + log(hnsw_ef_search))); (log(m) * (1 + log(ef)));
/* TODO incorporate ef_stream */
costs.numIndexTuples = (entryLevel * m) + costs.numIndexTuples = (entryLevel * m) +
(layer0TuplesMax * layer0Selectivity); (layer0TuplesMax * layer0Selectivity);
if (hnsw_streaming)
costs.numIndexTuples += EstimateExtraTuples(root, path, m, scalingFactor);
genericcostestimate(root, path, loop_count, &costs); genericcostestimate(root, path, loop_count, &costs);
get_tablespace_page_costs(path->indexinfo->reltablespace, NULL, &spc_seq_page_cost); get_tablespace_page_costs(path->indexinfo->reltablespace, NULL, &spc_seq_page_cost);