liuyu/pgvector
1
0
Fork 0
mirror of https://github.com/pgvector/pgvector.git synced 2026-07-12 23:56:55 +08:00
Code Issues Packages Projects Releases Wiki Activity

Update ivfflat cost estimation to choose index for more searches (#133)

Browse Source 
Co-authored-by: Andrew Kane <andrew@ankane.org>
This commit is contained in:
Jonathan S. Katz
2023-06-09 20:53:14 -04:00
committed by GitHub
parent 9f2359894f
commit 7446cbde8f
1 changed files with 38 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

47
src/ivfflat.c
View File

@@ -7,6 +7,7 @@
#include "ivfflat.h" #include "ivfflat.h"
#include "utils/guc.h" #include "utils/guc.h"
#include "utils/selfuncs.h" #include "utils/selfuncs.h"
#include "utils/spccache.h"
#if PG_VERSION_NUM >= 120000 #if PG_VERSION_NUM >= 120000
#include "commands/progress.h" #include "commands/progress.h"
@@ -69,7 +70,9 @@ ivfflatcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
GenericCosts costs; GenericCosts costs;
int lists; int lists;
double ratio; double ratio;
double spc_seq_page_cost;
Relation indexRel; Relation indexRel;
Cost startupCost;
#if PG_VERSION_NUM < 120000 #if PG_VERSION_NUM < 120000
List *qinfos; List *qinfos;
#endif #endif
@@ -87,6 +90,22 @@ ivfflatcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
MemSet(&costs, 0, sizeof(costs)); MemSet(&costs, 0, sizeof(costs));
indexRel = index_open(path->indexinfo->indexoid, NoLock);
lists = IvfflatGetLists(indexRel);
index_close(indexRel, NoLock);
/* Get the ratio of lists that we need to visit */
ratio = ((double) ivfflat_probes) / lists;
if (ratio > 1.0)
ratio = 1.0;
/*
* This gives us the subset of tuples to visit. This value is passed into
* the generic cost estimator to determine the number of pages to visit
* during the index scan.
*/
costs.numIndexTuples = path->indexinfo->rel->tuples * ratio;
#if PG_VERSION_NUM >= 120000 #if PG_VERSION_NUM >= 120000
genericcostestimate(root, path, loop_count, &costs); genericcostestimate(root, path, loop_count, &costs);
#else #else
@@ -94,18 +113,28 @@ ivfflatcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
genericcostestimate(root, path, loop_count, qinfos, &costs); genericcostestimate(root, path, loop_count, qinfos, &costs);
#endif #endif
indexRel = index_open(path->indexinfo->indexoid, NoLock); startupCost = costs.indexTotalCost;
lists = IvfflatGetLists(indexRel);
index_close(indexRel, NoLock);
ratio = ((double) ivfflat_probes) / lists; /* Adjust cost if needed since TOAST not included in seq scan cost */
if (ratio > 1) if (costs.numIndexPages > path->indexinfo->rel->pages && ratio < 0.5)
ratio = 1; {
get_tablespace_page_costs(path->indexinfo->reltablespace, NULL, &spc_seq_page_cost);
costs.indexTotalCost *= ratio; /* Change page cost from random to sequential */
startupCost -= costs.numIndexPages * (costs.spc_random_page_cost - spc_seq_page_cost);
/* Startup cost and total cost are same */ /* Remove cost of extra pages */
*indexStartupCost = costs.indexTotalCost; startupCost -= (costs.numIndexPages - path->indexinfo->rel->pages) * spc_seq_page_cost;
}
/*
* If the list selectivity is lower than what is returned from the generic
* cost estimator, use that.
*/
if (ratio < costs.indexSelectivity)
costs.indexSelectivity = ratio;
*indexStartupCost = startupCost;
*indexTotalCost = costs.indexTotalCost; *indexTotalCost = costs.indexTotalCost;
*indexSelectivity = costs.indexSelectivity; *indexSelectivity = costs.indexSelectivity;
*indexCorrelation = costs.indexCorrelation; *indexCorrelation = costs.indexCorrelation;