rdelaage
/
postgresql
mirror of https://git.postgresql.org/git/postgresql.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
postgresql/src/include/optimizer/cost.h

112 lines
4.1 KiB
C
Raw Normal View History

Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
/*-------------------------------------------------------------------------
*
Change my-function-name-- to my_function_name, and optimizer renames.
1999-02-14 00:22:53 +01:00
* cost.h
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
* prototypes for costsize.c and clausesel.c.
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
*
*
Update copyright for 2006. Update scripts.
2006-03-05 16:59:11 +01:00
* Portions Copyright (c) 1996-2006, PostgreSQL Global Development Group
Add: * Portions Copyright (c) 1996-2000, PostgreSQL, Inc to all files copyright Regents of Berkeley. Man, that's a lot of files.
2000-01-26 06:58:53 +01:00
* Portions Copyright (c) 1994, Regents of the University of California
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
*
Convert effective_cache_size to an integer, for better integration with upcoming units feature.
2006-07-26 13:35:56 +02:00
* $PostgreSQL: pgsql/src/include/optimizer/cost.h,v 1.78 2006/07/26 11:35:56 petere Exp $
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
*
*-------------------------------------------------------------------------
*/
#ifndef COST_H
#define COST_H
Finish implementation of hashed aggregation. Add enable_hashagg GUC parameter to allow it to be forced off for comparison purposes. Add ORDER BY clauses to a bunch of regression test queries that will otherwise produce randomly-ordered output in the new regime.
2002-11-21 01:42:20 +01:00
#include "nodes/plannodes.h"
Cleanup up include files.
1997-11-26 02:14:33 +01:00
#include "nodes/relation.h"
Finish implementation of hashed aggregation. Add enable_hashagg GUC parameter to allow it to be forced off for comparison purposes. Add ORDER BY clauses to a bunch of regression test queries that will otherwise produce randomly-ordered output in the new regime.
2002-11-21 01:42:20 +01:00
First cut at unifying regular selectivity estimation with indexscan selectivity estimation wasn't right. This is better...
2000-01-23 03:07:00 +01:00
/* defaults for costsize.c's Cost parameters */
New cost model for planning, incorporating a penalty for random page accesses versus sequential accesses, a (very crude) estimate of the effects of caching on random page accesses, and cost to evaluate WHERE- clause expressions. Export critical parameters for this model as SET variables. Also, create SET variables for the planner's enable flags (enable_seqscan, enable_indexscan, etc) so that these can be controlled more conveniently than via PGOPTIONS. Planner now estimates both startup cost (cost before retrieving first tuple) and total cost of each path, so it can optimize queries with LIMIT on a reasonable basis by interpolating between these costs. Same facility is a win for EXISTS(...) subqueries and some other cases. Redesign pathkey representation to achieve a major speedup in planning (I saw as much as 5X on a 10-way join); also minor changes in planner to reduce memory consumption by recycling discarded Path nodes and not constructing unnecessary lists. Minor cleanups to display more-plausible costs in some cases in EXPLAIN output. Initdb forced by change in interface to index cost estimation functions.
2000-02-15 21:49:31 +01:00
/* NB: cost-estimation code should use the variables, not these constants! */
Add all possible config file options.
2001-01-24 19:37:31 +01:00
/* If you change these, update backend/utils/misc/postgresql.sample.conf */
Add a GUC parameter seq_page_cost, and use that everywhere we formerly assumed that a sequential page fetch has cost 1.0. This patch doesn't in itself change the system's behavior at all, but it opens the door to people adopting other units of measurement for EXPLAIN costs. Also, if we ever decide it's worth inventing per-tablespace access cost settings, this change provides a workable intellectual framework for that.
2006-06-05 04:49:58 +02:00
#define DEFAULT_SEQ_PAGE_COST 1.0
New cost model for planning, incorporating a penalty for random page accesses versus sequential accesses, a (very crude) estimate of the effects of caching on random page accesses, and cost to evaluate WHERE- clause expressions. Export critical parameters for this model as SET variables. Also, create SET variables for the planner's enable flags (enable_seqscan, enable_indexscan, etc) so that these can be controlled more conveniently than via PGOPTIONS. Planner now estimates both startup cost (cost before retrieving first tuple) and total cost of each path, so it can optimize queries with LIMIT on a reasonable basis by interpolating between these costs. Same facility is a win for EXISTS(...) subqueries and some other cases. Redesign pathkey representation to achieve a major speedup in planning (I saw as much as 5X on a 10-way join); also minor changes in planner to reduce memory consumption by recycling discarded Path nodes and not constructing unnecessary lists. Minor cleanups to display more-plausible costs in some cases in EXPLAIN output. Initdb forced by change in interface to index cost estimation functions.
2000-02-15 21:49:31 +01:00
#define DEFAULT_RANDOM_PAGE_COST 4.0
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
#define DEFAULT_CPU_TUPLE_COST 0.01
Increase the default value of cpu_index_tuple_cost from 0.001 to 0.005. This shouldn't affect simple indexscans much, while for bitmap scans that are touching a lot of index rows, this seems to bring the estimates more in line with reality. Per recent discussion.
2006-06-05 05:03:42 +02:00
#define DEFAULT_CPU_INDEX_TUPLE_COST 0.005
New cost model for planning, incorporating a penalty for random page accesses versus sequential accesses, a (very crude) estimate of the effects of caching on random page accesses, and cost to evaluate WHERE- clause expressions. Export critical parameters for this model as SET variables. Also, create SET variables for the planner's enable flags (enable_seqscan, enable_indexscan, etc) so that these can be controlled more conveniently than via PGOPTIONS. Planner now estimates both startup cost (cost before retrieving first tuple) and total cost of each path, so it can optimize queries with LIMIT on a reasonable basis by interpolating between these costs. Same facility is a win for EXISTS(...) subqueries and some other cases. Redesign pathkey representation to achieve a major speedup in planning (I saw as much as 5X on a 10-way join); also minor changes in planner to reduce memory consumption by recycling discarded Path nodes and not constructing unnecessary lists. Minor cleanups to display more-plausible costs in some cases in EXPLAIN output. Initdb forced by change in interface to index cost estimation functions.
2000-02-15 21:49:31 +01:00
#define DEFAULT_CPU_OPERATOR_COST 0.0025
First cut at unifying regular selectivity estimation with indexscan selectivity estimation wasn't right. This is better...
2000-01-23 03:07:00 +01:00
Convert effective_cache_size to an integer, for better integration with upcoming units feature.
2006-07-26 13:35:56 +02:00
#define DEFAULT_EFFECTIVE_CACHE_SIZE 1000 /* measured in pages */
Add a GUC parameter seq_page_cost, and use that everywhere we formerly assumed that a sequential page fetch has cost 1.0. This patch doesn't in itself change the system's behavior at all, but it opens the door to people adopting other units of measurement for EXPLAIN costs. Also, if we ever decide it's worth inventing per-tablespace access cost settings, this change provides a workable intellectual framework for that.
2006-06-05 04:49:58 +02:00
Clean up #include in /include directory. Add scripts for checking includes.
1999-07-15 17:21:54 +02:00
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
/*
Change my-function-name-- to my_function_name, and optimizer renames.
1999-02-14 00:22:53 +01:00
* prototypes for costsize.c
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
* routines to compute costs and sizes
*/
Revise handling of index-type-specific indexscan cost estimation, per pghackers discussion of 5-Jan-2000. The amopselect and amopnpages estimators are gone, and in their place is a per-AM amcostestimate procedure (linked to from pg_am, not pg_amop).
2000-01-23 00:50:30 +01:00
New cost model for planning, incorporating a penalty for random page accesses versus sequential accesses, a (very crude) estimate of the effects of caching on random page accesses, and cost to evaluate WHERE- clause expressions. Export critical parameters for this model as SET variables. Also, create SET variables for the planner's enable flags (enable_seqscan, enable_indexscan, etc) so that these can be controlled more conveniently than via PGOPTIONS. Planner now estimates both startup cost (cost before retrieving first tuple) and total cost of each path, so it can optimize queries with LIMIT on a reasonable basis by interpolating between these costs. Same facility is a win for EXISTS(...) subqueries and some other cases. Redesign pathkey representation to achieve a major speedup in planning (I saw as much as 5X on a 10-way join); also minor changes in planner to reduce memory consumption by recycling discarded Path nodes and not constructing unnecessary lists. Minor cleanups to display more-plausible costs in some cases in EXPLAIN output. Initdb forced by change in interface to index cost estimation functions.
2000-02-15 21:49:31 +01:00
/* parameter variables and flags */
Add a GUC parameter seq_page_cost, and use that everywhere we formerly assumed that a sequential page fetch has cost 1.0. This patch doesn't in itself change the system's behavior at all, but it opens the door to people adopting other units of measurement for EXPLAIN costs. Also, if we ever decide it's worth inventing per-tablespace access cost settings, this change provides a workable intellectual framework for that.
2006-06-05 04:49:58 +02:00
extern DLLIMPORT double seq_page_cost;
extern DLLIMPORT double random_page_cost;
extern DLLIMPORT double cpu_tuple_cost;
Add missing DLLIMPORT for cpu_index_tuple_cost to src/include/optimizer/cost.h. This is required to compile the PostGIS extension module with Cygwin http://postgis.refractions.net Norman Vine
2003-06-11 17:01:15 +02:00
extern DLLIMPORT double cpu_index_tuple_cost;
Add a GUC parameter seq_page_cost, and use that everywhere we formerly assumed that a sequential page fetch has cost 1.0. This patch doesn't in itself change the system's behavior at all, but it opens the door to people adopting other units of measurement for EXPLAIN costs. Also, if we ever decide it's worth inventing per-tablespace access cost settings, this change provides a workable intellectual framework for that.
2006-06-05 04:49:58 +02:00
extern DLLIMPORT double cpu_operator_cost;
Convert effective_cache_size to an integer, for better integration with upcoming units feature.
2006-07-26 13:35:56 +02:00
extern DLLIMPORT int effective_cache_size;
Revise handling of index-type-specific indexscan cost estimation, per pghackers discussion of 5-Jan-2000. The amopselect and amopnpages estimators are gone, and in their place is a per-AM amcostestimate procedure (linked to from pg_am, not pg_amop).
2000-01-23 00:50:30 +01:00
extern Cost disable_cost;
extern bool enable_seqscan;
extern bool enable_indexscan;
Rethink original decision to use AND/OR Expr nodes to represent bitmap logic operations during planning. Seems cleaner to create two new Path node types, instead --- this avoids duplication of cost-estimation code. Also, create an enable_bitmapscan GUC parameter to control use of bitmap plans.
2005-04-21 21:18:13 +02:00
extern bool enable_bitmapscan;
Revise handling of index-type-specific indexscan cost estimation, per pghackers discussion of 5-Jan-2000. The amopselect and amopnpages estimators are gone, and in their place is a per-AM amcostestimate procedure (linked to from pg_am, not pg_amop).
2000-01-23 00:50:30 +01:00
extern bool enable_tidscan;
extern bool enable_sort;
Finish implementation of hashed aggregation. Add enable_hashagg GUC parameter to allow it to be forced off for comparison purposes. Add ORDER BY clauses to a bunch of regression test queries that will otherwise produce randomly-ordered output in the new regime.
2002-11-21 01:42:20 +01:00
extern bool enable_hashagg;
Revise handling of index-type-specific indexscan cost estimation, per pghackers discussion of 5-Jan-2000. The amopselect and amopnpages estimators are gone, and in their place is a per-AM amcostestimate procedure (linked to from pg_am, not pg_amop).
2000-01-23 00:50:30 +01:00
extern bool enable_nestloop;
extern bool enable_mergejoin;
extern bool enable_hashjoin;
enable_constraint_exclusion => constraint_exclusion Also improve wording.
2005-08-22 19:35:03 +02:00
extern bool constraint_exclusion;
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
Add the ability to extract OR indexscan conditions from OR-of-AND join conditions in which each OR subclause includes a constraint on the same relation. This implements the other useful side-effect of conversion to CNF format, without its unpleasant side-effects. As per pghackers discussion of a few weeks ago.
2004-01-05 06:07:36 +01:00
extern double clamp_row_est(double nrows);
Make the planner estimate costs for nestloop inner indexscans on the basis that the Mackert-Lohmann formula applies across all the repetitions of the nestloop, not just each scan independently. We use the M-L formula to estimate the number of pages fetched from the index as well as from the table; that isn't what it was designed for, but it seems reasonably applicable anyway. This makes large numbers of repetitions look much cheaper than before, which accords with many reports we've received of overestimation of the cost of a nestloop. Also, change the index access cost model to charge random_page_cost per index leaf page touched, while explicitly not counting anything for access to metapage or upper tree pages. This may all need tweaking after we get some field experience, but in simple tests it seems to be giving saner results than before. The main thing is to get the infrastructure in place to let cost_index() and amcostestimate functions take repeated scans into account at all. Per my recent proposal. Note: this patch changes pg_proc.h, but I did not force initdb because the changes are basically cosmetic --- the system does not look into pg_proc to decide how to call an index amcostestimate function, and there's no way to call such a function from SQL at all.
2006-06-06 19:59:58 +02:00
extern double index_pages_fetched(double tuples_fetched, BlockNumber pages,
BlockNumber other_pages);
Remove planner's private fields from Query struct, and put them into a new PlannerInfo struct, which is passed around instead of the bare Query in all the planning code. This commit is essentially just a code-beautification exercise, but it does open the door to making larger changes to the planner data structures without having to muck with the widely-known Query struct.
2005-06-06 00:32:58 +02:00
extern void cost_seqscan(Path *path, PlannerInfo *root, RelOptInfo *baserel);
extern void cost_index(IndexPath *path, PlannerInfo *root, IndexOptInfo *index,
Make the planner estimate costs for nestloop inner indexscans on the basis that the Mackert-Lohmann formula applies across all the repetitions of the nestloop, not just each scan independently. We use the M-L formula to estimate the number of pages fetched from the index as well as from the table; that isn't what it was designed for, but it seems reasonably applicable anyway. This makes large numbers of repetitions look much cheaper than before, which accords with many reports we've received of overestimation of the cost of a nestloop. Also, change the index access cost model to charge random_page_cost per index leaf page touched, while explicitly not counting anything for access to metapage or upper tree pages. This may all need tweaking after we get some field experience, but in simple tests it seems to be giving saner results than before. The main thing is to get the infrastructure in place to let cost_index() and amcostestimate functions take repeated scans into account at all. Per my recent proposal. Note: this patch changes pg_proc.h, but I did not force initdb because the changes are basically cosmetic --- the system does not look into pg_proc to decide how to call an index amcostestimate function, and there's no way to call such a function from SQL at all.
2006-06-06 19:59:58 +02:00
List *indexQuals, RelOptInfo *outer_rel);
Remove planner's private fields from Query struct, and put them into a new PlannerInfo struct, which is passed around instead of the bare Query in all the planning code. This commit is essentially just a code-beautification exercise, but it does open the door to making larger changes to the planner data structures without having to muck with the widely-known Query struct.
2005-06-06 00:32:58 +02:00
extern void cost_bitmap_heap_scan(Path *path, PlannerInfo *root, RelOptInfo *baserel,
In the recent changes to make the planner account better for cache effects in a nestloop inner indexscan, I had only dealt with plain index scans and the index portion of bitmap scans. But there will be cache benefits for the heap accesses of bitmap scans too, so fix cost_bitmap_heap_scan() to account for that.
2006-07-22 17:41:56 +02:00
Path *bitmapqual, RelOptInfo *outer_rel);
Remove planner's private fields from Query struct, and put them into a new PlannerInfo struct, which is passed around instead of the bare Query in all the planning code. This commit is essentially just a code-beautification exercise, but it does open the door to making larger changes to the planner data structures without having to muck with the widely-known Query struct.
2005-06-06 00:32:58 +02:00
extern void cost_bitmap_and_node(BitmapAndPath *path, PlannerInfo *root);
extern void cost_bitmap_or_node(BitmapOrPath *path, PlannerInfo *root);
First cut at planner support for bitmap index scans. Lots to do yet, but the code is basically working. Along the way, rewrite the entire approach to processing OR index conditions, and make it work in join cases for the first time ever. orindxpath.c is now basically obsolete, but I left it in for the time being to allow easy comparison testing against the old implementation.
2005-04-22 23:58:32 +02:00
extern void cost_bitmap_tree_node(Path *path, Cost *cost, Selectivity *selec);
Remove planner's private fields from Query struct, and put them into a new PlannerInfo struct, which is passed around instead of the bare Query in all the planning code. This commit is essentially just a code-beautification exercise, but it does open the door to making larger changes to the planner data structures without having to muck with the widely-known Query struct.
2005-06-06 00:32:58 +02:00
extern void cost_tidscan(Path *path, PlannerInfo *root,
Teach tid-scan code to make use of "ctid = ANY (array)" clauses, so that "ctid IN (list)" will still work after we convert IN to ScalarArrayOpExpr. Make some minor efficiency improvements while at it, such as ensuring that multiple TIDs are fetched in physical heap order. And fix EXPLAIN so that it shows what's really going on for a TID scan.
2005-11-26 23:14:57 +01:00
RelOptInfo *baserel, List *tidquals);
Make cost estimates for SubqueryScan more realistic: charge cpu_tuple_cost for each row processed, and don't forget the evaluation cost of any restriction clauses attached to the node. Per discussion with Greg Stark.
2003-07-15 00:35:54 +02:00
extern void cost_subqueryscan(Path *path, RelOptInfo *baserel);
Remove planner's private fields from Query struct, and put them into a new PlannerInfo struct, which is passed around instead of the bare Query in all the planning code. This commit is essentially just a code-beautification exercise, but it does open the door to making larger changes to the planner data structures without having to muck with the widely-known Query struct.
2005-06-06 00:32:58 +02:00
extern void cost_functionscan(Path *path, PlannerInfo *root,
pgindent run.
2002-09-04 22:31:48 +02:00
RelOptInfo *baserel);
Remove planner's private fields from Query struct, and put them into a new PlannerInfo struct, which is passed around instead of the bare Query in all the planning code. This commit is essentially just a code-beautification exercise, but it does open the door to making larger changes to the planner data structures without having to muck with the widely-known Query struct.
2005-06-06 00:32:58 +02:00
extern void cost_sort(Path *path, PlannerInfo *root,
Finish implementation of hashed aggregation. Add enable_hashagg GUC parameter to allow it to be forced off for comparison purposes. Add ORDER BY clauses to a bunch of regression test queries that will otherwise produce randomly-ordered output in the new regime.
2002-11-21 01:42:20 +01:00
List *pathkeys, Cost input_cost, double tuples, int width);
Be more realistic about plans involving Materialize nodes: take their cost into account while planning.
2002-11-30 06:21:03 +01:00
extern void cost_material(Path *path,
pgindent run.
2003-08-04 02:43:34 +02:00
Cost input_cost, double tuples, int width);
Remove planner's private fields from Query struct, and put them into a new PlannerInfo struct, which is passed around instead of the bare Query in all the planning code. This commit is essentially just a code-beautification exercise, but it does open the door to making larger changes to the planner data structures without having to muck with the widely-known Query struct.
2005-06-06 00:32:58 +02:00
extern void cost_agg(Path *path, PlannerInfo *root,
pgindent run.
2003-08-04 02:43:34 +02:00
AggStrategy aggstrategy, int numAggs,
int numGroupCols, double numGroups,
Cost input_startup_cost, Cost input_total_cost,
double input_tuples);
Remove planner's private fields from Query struct, and put them into a new PlannerInfo struct, which is passed around instead of the bare Query in all the planning code. This commit is essentially just a code-beautification exercise, but it does open the door to making larger changes to the planner data structures without having to muck with the widely-known Query struct.
2005-06-06 00:32:58 +02:00
extern void cost_group(Path *path, PlannerInfo *root,
pgindent run.
2003-08-04 02:43:34 +02:00
int numGroupCols, double numGroups,
Cost input_startup_cost, Cost input_total_cost,
double input_tuples);
Remove planner's private fields from Query struct, and put them into a new PlannerInfo struct, which is passed around instead of the bare Query in all the planning code. This commit is essentially just a code-beautification exercise, but it does open the door to making larger changes to the planner data structures without having to muck with the widely-known Query struct.
2005-06-06 00:32:58 +02:00
extern void cost_nestloop(NestPath *path, PlannerInfo *root);
extern void cost_mergejoin(MergePath *path, PlannerInfo *root);
extern void cost_hashjoin(HashPath *path, PlannerInfo *root);
Another pgindent run with updated typedefs.
2003-08-08 23:42:59 +02:00
extern void cost_qual_eval(QualCost *cost, List *quals);
Remove planner's private fields from Query struct, and put them into a new PlannerInfo struct, which is passed around instead of the bare Query in all the planning code. This commit is essentially just a code-beautification exercise, but it does open the door to making larger changes to the planner data structures without having to muck with the widely-known Query struct.
2005-06-06 00:32:58 +02:00
extern void set_baserel_size_estimates(PlannerInfo *root, RelOptInfo *rel);
extern void set_joinrel_size_estimates(PlannerInfo *root, RelOptInfo *rel,
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
RelOptInfo *outer_rel,
RelOptInfo *inner_rel,
Take OUTER JOIN semantics into account when estimating the size of join relations. It's not very bright, but at least it now knows that A LEFT JOIN B must produce at least as many rows as are in A ...
2001-02-16 01:03:08 +01:00
JoinType jointype,
Ye-old pgindent run. Same 4-space tabs.
2000-04-12 19:17:23 +02:00
List *restrictlist);
Remove planner's private fields from Query struct, and put them into a new PlannerInfo struct, which is passed around instead of the bare Query in all the planning code. This commit is essentially just a code-beautification exercise, but it does open the door to making larger changes to the planner data structures without having to muck with the widely-known Query struct.
2005-06-06 00:32:58 +02:00
extern void set_function_size_estimates(PlannerInfo *root, RelOptInfo *rel);
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
/*
Another round of planner/optimizer work. This is just restructuring and code cleanup; no major improvements yet. However, EXPLAIN does produce more intuitive outputs for nested loops with indexscans now...
2000-01-09 01:26:47 +01:00
* prototypes for clausesel.c
Massive commit to run PGINDENT on all *.c and *.h files.
1997-09-07 07:04:48 +02:00
* routines to compute clause selectivities
Postgres95 1.01 Distribution - Virgin Sources
1996-07-09 08:22:35 +02:00
*/
Remove planner's private fields from Query struct, and put them into a new PlannerInfo struct, which is passed around instead of the bare Query in all the planning code. This commit is essentially just a code-beautification exercise, but it does open the door to making larger changes to the planner data structures without having to muck with the widely-known Query struct.
2005-06-06 00:32:58 +02:00
extern Selectivity clauselist_selectivity(PlannerInfo *root,
pgindent run.
2003-08-04 02:43:34 +02:00
List *clauses,
int varRelid,
JoinType jointype);
Remove planner's private fields from Query struct, and put them into a new PlannerInfo struct, which is passed around instead of the bare Query in all the planning code. This commit is essentially just a code-beautification exercise, but it does open the door to making larger changes to the planner data structures without having to muck with the widely-known Query struct.
2005-06-06 00:32:58 +02:00
extern Selectivity clause_selectivity(PlannerInfo *root,
pgindent run.
2003-08-04 02:43:34 +02:00
Node *clause,
int varRelid,
JoinType jointype);
Another pgindent run. Fixes enum indenting, and improves #endif spacing. Also adds space for one-line comments.
2001-10-28 07:26:15 +01:00
New pgindent run with fixes suggested by Tom. Patch manually reviewed, initdb/regression tests pass.
2001-11-05 18:46:40 +01:00
#endif /* COST_H */