database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-08-12 15:23:02 +03:00
Code Activity

Reduce the size of memory allocations by lazy vacuum when processing a small

Browse Source 
table, by allocating just enough for a hardcoded number of dead tuples per
page.  The current estimate is 200 dead tuples per page.

Per reports from Jeff Amiel, Erik Jones and Marko Kreen, and subsequent
discussion.
CVS: ----------------------------------------------------------------------
CVS: Enter Log.  Lines beginning with `CVS:' are removed automatically
CVS:
CVS: Committing in .
CVS:
CVS: Modified Files:
CVS: 	commands/vacuumlazy.c
CVS: ----------------------------------------------------------------------
This commit is contained in:
Alvaro Herrera
2007-09-24 03:53:12 +00:00
parent f0a06adfe7
commit 46dcd202ef
1 changed files with 17 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
src/backend/commands/vacuumlazy.c
View File

@@ -11,10 +11,12 @@
* on the number of tuples and pages we will keep track of at once. * on the number of tuples and pages we will keep track of at once.
* *
* We are willing to use at most maintenance_work_mem memory space to keep * We are willing to use at most maintenance_work_mem memory space to keep
* track of dead tuples. We initially allocate an array of TIDs of that size. * track of dead tuples. We initially allocate an array of TIDs of that size,
* If the array threatens to overflow, we suspend the heap scan phase and * with an upper limit that depends on table size (this limit ensures we don't
* perform a pass of index cleanup and page compaction, then resume the heap * allocate a huge area uselessly for vacuuming small tables). If the array
* scan with an empty TID array. * threatens to overflow, we suspend the heap scan phase and perform a pass of
* index cleanup and page compaction, then resume the heap scan with an empty
* TID array.
* *
* We can limit the storage for page free space to MaxFSMPages entries, * We can limit the storage for page free space to MaxFSMPages entries,
* since that's the most the free space map will be willing to remember * since that's the most the free space map will be willing to remember
@@ -31,7 +33,7 @@
* *
* *
* IDENTIFICATION * IDENTIFICATION
* $PostgreSQL: pgsql/src/backend/commands/vacuumlazy.c,v 1.50.4.5 2007/09/16 02:38:14 tgl Exp $ * $PostgreSQL: pgsql/src/backend/commands/vacuumlazy.c,v 1.50.4.6 2007/09/24 03:53:12 alvherre Exp $
* *
*------------------------------------------------------------------------- *-------------------------------------------------------------------------
*/ */
@@ -63,6 +65,12 @@
/* MAX_TUPLES_PER_PAGE can be a conservative upper limit */ /* MAX_TUPLES_PER_PAGE can be a conservative upper limit */
#define MAX_TUPLES_PER_PAGE ((int) (BLCKSZ / sizeof(HeapTupleHeaderData))) #define MAX_TUPLES_PER_PAGE ((int) (BLCKSZ / sizeof(HeapTupleHeaderData)))
/*
* Guesstimation of number of dead tuples per page. This is used to
* provide an upper limit to memory allocated when vacuuming small
* tables.
*/
#define LAZY_ALLOC_TUPLES 200
typedef struct LVRelStats typedef struct LVRelStats
{ {
@@ -942,6 +950,10 @@ lazy_space_alloc(LVRelStats *vacrelstats, BlockNumber relblocks)
if (maxtuples < MAX_TUPLES_PER_PAGE) if (maxtuples < MAX_TUPLES_PER_PAGE)
maxtuples = MAX_TUPLES_PER_PAGE; maxtuples = MAX_TUPLES_PER_PAGE;
/* curious coding here to ensure the multiplication can't overflow */
if ((BlockNumber) (maxtuples / LAZY_ALLOC_TUPLES) > relblocks)
maxtuples = relblocks * LAZY_ALLOC_TUPLES;
vacrelstats->num_dead_tuples = 0; vacrelstats->num_dead_tuples = 0;
vacrelstats->max_dead_tuples = maxtuples; vacrelstats->max_dead_tuples = maxtuples;
vacrelstats->dead_tuples = (ItemPointer) vacrelstats->dead_tuples = (ItemPointer)