postgresql/src/backend/access/nbtree/nbtree.c

585 lines
16 KiB
C
Raw Normal View History

/*-------------------------------------------------------------------------
*
* btree.c--
* Implementation of Lehman and Yao's btree management algorithm for
* Postgres.
*
* Copyright (c) 1994, Regents of the University of California
*
*
* IDENTIFICATION
* $Header: /cvsroot/pgsql/src/backend/access/nbtree/nbtree.c,v 1.17 1997/03/24 08:48:11 vadim Exp $
*
* NOTES
* This file contains only the public interface routines.
*
*-------------------------------------------------------------------------
*/
1996-11-05 11:35:38 +01:00
#include <postgres.h>
#include <access/genam.h>
#include <storage/bufpage.h>
#include <storage/bufmgr.h>
#include <access/nbtree.h>
#include <executor/executor.h>
#include <access/heapam.h>
#include <catalog/index.h>
#include <miscadmin.h>
#ifndef HAVE_MEMMOVE
# include <regex/utils.h>
#else
# include <string.h>
#endif
What looks like some *major* improvements to btree indexing... Patches from: aoki@CS.Berkeley.EDU (Paul M. Aoki) i gave jolly my btree bulkload code a long, long time ago but never gave him a bunch of my bugfixes. here's a diff against the 6.0 baseline. for some reason, this code has slowed down somewhat relative to the insertion-build code on very small tables. don't know why -- it used to be within about 10%. anyway, here are some (highly unscientific!) timings on a dec 3000/300 for synthetic tables with 10k, 100k and 1000k tuples (basically, 1mb, 10mb and 100mb heaps). 'c' means clustered (pre-sorted) inputs and 'u' means unclustered (randomly ordered) inputs. the 10k table basically fits in the buffer pool, but the 100k and 1000k tables don't. as you can see, insertion build is fine if you've sorted your heaps on your index key or if your heap fits in core, but is absolutely horrible on unordered data (yes, that's 7.5 hours to index 100mb of data...) because of the zillions of random i/os. if it doesn't work for you for whatever reason, you can always turn it back off by flipping the FastBuild flag in nbtree.c. i don't have time to maintain it. good luck! baseline code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 8.6 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 9.1 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.2 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 652.4 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.1 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 26772.9 bulkloading code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 11.3 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 10.4 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.5 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 63.5 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.9 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 701.0
1997-02-12 06:04:52 +01:00
bool BuildingBtree = false; /* see comment in btbuild() */
bool FastBuild = true; /* use sort/build instead of insertion build */
/*
* btbuild() -- build a new btree index.
*
* We use a global variable to record the fact that we're creating
* a new index. This is used to avoid high-concurrency locking,
* since the index won't be visible until this transaction commits
* and since building is guaranteed to be single-threaded.
*/
void
btbuild(Relation heap,
Relation index,
int natts,
AttrNumber *attnum,
IndexStrategy istrat,
uint16 pcount,
Datum *params,
FuncIndexInfo *finfo,
PredInfo *predInfo)
{
HeapScanDesc hscan;
Buffer buffer;
HeapTuple htup;
IndexTuple itup;
TupleDesc htupdesc, itupdesc;
Datum *attdata;
bool *nulls;
InsertIndexResult res = 0;
int nhtups, nitups;
int i;
BTItem btitem;
#ifndef OMIT_PARTIAL_INDEX
What looks like some *major* improvements to btree indexing... Patches from: aoki@CS.Berkeley.EDU (Paul M. Aoki) i gave jolly my btree bulkload code a long, long time ago but never gave him a bunch of my bugfixes. here's a diff against the 6.0 baseline. for some reason, this code has slowed down somewhat relative to the insertion-build code on very small tables. don't know why -- it used to be within about 10%. anyway, here are some (highly unscientific!) timings on a dec 3000/300 for synthetic tables with 10k, 100k and 1000k tuples (basically, 1mb, 10mb and 100mb heaps). 'c' means clustered (pre-sorted) inputs and 'u' means unclustered (randomly ordered) inputs. the 10k table basically fits in the buffer pool, but the 100k and 1000k tables don't. as you can see, insertion build is fine if you've sorted your heaps on your index key or if your heap fits in core, but is absolutely horrible on unordered data (yes, that's 7.5 hours to index 100mb of data...) because of the zillions of random i/os. if it doesn't work for you for whatever reason, you can always turn it back off by flipping the FastBuild flag in nbtree.c. i don't have time to maintain it. good luck! baseline code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 8.6 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 9.1 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.2 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 652.4 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.1 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 26772.9 bulkloading code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 11.3 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 10.4 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.5 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 63.5 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.9 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 701.0
1997-02-12 06:04:52 +01:00
ExprContext *econtext = (ExprContext *) NULL;
TupleTable tupleTable = (TupleTable) NULL;
TupleTableSlot *slot = (TupleTableSlot *) NULL;
#endif
Oid hrelid, irelid;
Node *pred, *oldPred;
What looks like some *major* improvements to btree indexing... Patches from: aoki@CS.Berkeley.EDU (Paul M. Aoki) i gave jolly my btree bulkload code a long, long time ago but never gave him a bunch of my bugfixes. here's a diff against the 6.0 baseline. for some reason, this code has slowed down somewhat relative to the insertion-build code on very small tables. don't know why -- it used to be within about 10%. anyway, here are some (highly unscientific!) timings on a dec 3000/300 for synthetic tables with 10k, 100k and 1000k tuples (basically, 1mb, 10mb and 100mb heaps). 'c' means clustered (pre-sorted) inputs and 'u' means unclustered (randomly ordered) inputs. the 10k table basically fits in the buffer pool, but the 100k and 1000k tables don't. as you can see, insertion build is fine if you've sorted your heaps on your index key or if your heap fits in core, but is absolutely horrible on unordered data (yes, that's 7.5 hours to index 100mb of data...) because of the zillions of random i/os. if it doesn't work for you for whatever reason, you can always turn it back off by flipping the FastBuild flag in nbtree.c. i don't have time to maintain it. good luck! baseline code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 8.6 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 9.1 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.2 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 652.4 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.1 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 26772.9 bulkloading code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 11.3 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 10.4 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.5 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 63.5 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.9 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 701.0
1997-02-12 06:04:52 +01:00
void *spool = (void *) NULL;
bool isunique;
What looks like some *major* improvements to btree indexing... Patches from: aoki@CS.Berkeley.EDU (Paul M. Aoki) i gave jolly my btree bulkload code a long, long time ago but never gave him a bunch of my bugfixes. here's a diff against the 6.0 baseline. for some reason, this code has slowed down somewhat relative to the insertion-build code on very small tables. don't know why -- it used to be within about 10%. anyway, here are some (highly unscientific!) timings on a dec 3000/300 for synthetic tables with 10k, 100k and 1000k tuples (basically, 1mb, 10mb and 100mb heaps). 'c' means clustered (pre-sorted) inputs and 'u' means unclustered (randomly ordered) inputs. the 10k table basically fits in the buffer pool, but the 100k and 1000k tables don't. as you can see, insertion build is fine if you've sorted your heaps on your index key or if your heap fits in core, but is absolutely horrible on unordered data (yes, that's 7.5 hours to index 100mb of data...) because of the zillions of random i/os. if it doesn't work for you for whatever reason, you can always turn it back off by flipping the FastBuild flag in nbtree.c. i don't have time to maintain it. good luck! baseline code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 8.6 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 9.1 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.2 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 652.4 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.1 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 26772.9 bulkloading code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 11.3 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 10.4 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.5 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 63.5 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.9 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 701.0
1997-02-12 06:04:52 +01:00
bool usefast;
#if 0
ResetBufferUsage();
#endif
/* note that this is a new btree */
BuildingBtree = true;
pred = predInfo->pred;
oldPred = predInfo->oldPred;
What looks like some *major* improvements to btree indexing... Patches from: aoki@CS.Berkeley.EDU (Paul M. Aoki) i gave jolly my btree bulkload code a long, long time ago but never gave him a bunch of my bugfixes. here's a diff against the 6.0 baseline. for some reason, this code has slowed down somewhat relative to the insertion-build code on very small tables. don't know why -- it used to be within about 10%. anyway, here are some (highly unscientific!) timings on a dec 3000/300 for synthetic tables with 10k, 100k and 1000k tuples (basically, 1mb, 10mb and 100mb heaps). 'c' means clustered (pre-sorted) inputs and 'u' means unclustered (randomly ordered) inputs. the 10k table basically fits in the buffer pool, but the 100k and 1000k tables don't. as you can see, insertion build is fine if you've sorted your heaps on your index key or if your heap fits in core, but is absolutely horrible on unordered data (yes, that's 7.5 hours to index 100mb of data...) because of the zillions of random i/os. if it doesn't work for you for whatever reason, you can always turn it back off by flipping the FastBuild flag in nbtree.c. i don't have time to maintain it. good luck! baseline code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 8.6 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 9.1 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.2 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 652.4 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.1 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 26772.9 bulkloading code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 11.3 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 10.4 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.5 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 63.5 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.9 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 701.0
1997-02-12 06:04:52 +01:00
/*
* bootstrap processing does something strange, so don't use
* sort/build for initial catalog indices. at some point i need
* to look harder at this. (there is some kind of incremental
* processing going on there.) -- pma 08/29/95
*/
usefast = (FastBuild && IsNormalProcessingMode());
/* see if index is unique */
isunique = IndexIsUniqueNoCache(RelationGetRelationId(index));
/* initialize the btree index metadata page (if this is a new index) */
if (oldPred == NULL)
_bt_metapinit(index);
/* get tuple descriptors for heap and index relations */
htupdesc = RelationGetTupleDescriptor(heap);
itupdesc = RelationGetTupleDescriptor(index);
/* get space for data items that'll appear in the index tuple */
attdata = (Datum *) palloc(natts * sizeof(Datum));
nulls = (bool *) palloc(natts * sizeof(bool));
/*
* If this is a predicate (partial) index, we will need to evaluate the
* predicate using ExecQual, which requires the current tuple to be in a
* slot of a TupleTable. In addition, ExecQual must have an ExprContext
* referring to that slot. Here, we initialize dummy TupleTable and
* ExprContext objects for this purpose. --Nels, Feb '92
*/
#ifndef OMIT_PARTIAL_INDEX
if (pred != NULL || oldPred != NULL) {
tupleTable = ExecCreateTupleTable(1);
slot = ExecAllocTableSlot(tupleTable);
econtext = makeNode(ExprContext);
FillDummyExprContext(econtext, slot, htupdesc, InvalidBuffer);
What looks like some *major* improvements to btree indexing... Patches from: aoki@CS.Berkeley.EDU (Paul M. Aoki) i gave jolly my btree bulkload code a long, long time ago but never gave him a bunch of my bugfixes. here's a diff against the 6.0 baseline. for some reason, this code has slowed down somewhat relative to the insertion-build code on very small tables. don't know why -- it used to be within about 10%. anyway, here are some (highly unscientific!) timings on a dec 3000/300 for synthetic tables with 10k, 100k and 1000k tuples (basically, 1mb, 10mb and 100mb heaps). 'c' means clustered (pre-sorted) inputs and 'u' means unclustered (randomly ordered) inputs. the 10k table basically fits in the buffer pool, but the 100k and 1000k tables don't. as you can see, insertion build is fine if you've sorted your heaps on your index key or if your heap fits in core, but is absolutely horrible on unordered data (yes, that's 7.5 hours to index 100mb of data...) because of the zillions of random i/os. if it doesn't work for you for whatever reason, you can always turn it back off by flipping the FastBuild flag in nbtree.c. i don't have time to maintain it. good luck! baseline code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 8.6 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 9.1 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.2 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 652.4 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.1 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 26772.9 bulkloading code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 11.3 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 10.4 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.5 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 63.5 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.9 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 701.0
1997-02-12 06:04:52 +01:00
/*
* we never want to use sort/build if we are extending an
* existing partial index -- it works by inserting the
* newly-qualifying tuples into the existing index.
* (sort/build would overwrite the existing index with one
* consisting of the newly-qualifying tuples.)
*/
usefast = false;
}
#endif /* OMIT_PARTIAL_INDEX */
/* start a heap scan */
hscan = heap_beginscan(heap, 0, NowTimeQual, 0, (ScanKey) NULL);
htup = heap_getnext(hscan, 0, &buffer);
/* build the index */
nhtups = nitups = 0;
What looks like some *major* improvements to btree indexing... Patches from: aoki@CS.Berkeley.EDU (Paul M. Aoki) i gave jolly my btree bulkload code a long, long time ago but never gave him a bunch of my bugfixes. here's a diff against the 6.0 baseline. for some reason, this code has slowed down somewhat relative to the insertion-build code on very small tables. don't know why -- it used to be within about 10%. anyway, here are some (highly unscientific!) timings on a dec 3000/300 for synthetic tables with 10k, 100k and 1000k tuples (basically, 1mb, 10mb and 100mb heaps). 'c' means clustered (pre-sorted) inputs and 'u' means unclustered (randomly ordered) inputs. the 10k table basically fits in the buffer pool, but the 100k and 1000k tables don't. as you can see, insertion build is fine if you've sorted your heaps on your index key or if your heap fits in core, but is absolutely horrible on unordered data (yes, that's 7.5 hours to index 100mb of data...) because of the zillions of random i/os. if it doesn't work for you for whatever reason, you can always turn it back off by flipping the FastBuild flag in nbtree.c. i don't have time to maintain it. good luck! baseline code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 8.6 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 9.1 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.2 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 652.4 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.1 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 26772.9 bulkloading code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 11.3 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 10.4 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.5 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 63.5 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.9 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 701.0
1997-02-12 06:04:52 +01:00
if (usefast) {
spool = _bt_spoolinit(index, 7, isunique);
res = (InsertIndexResult) NULL;
}
for (; HeapTupleIsValid(htup); htup = heap_getnext(hscan, 0, &buffer)) {
nhtups++;
/*
* If oldPred != NULL, this is an EXTEND INDEX command, so skip
* this tuple if it was already in the existing partial index
*/
if (oldPred != NULL) {
#ifndef OMIT_PARTIAL_INDEX
/*SetSlotContents(slot, htup);*/
slot->val = htup;
if (ExecQual((List*)oldPred, econtext) == true) {
nitups++;
continue;
}
#endif /* OMIT_PARTIAL_INDEX */
}
/* Skip this tuple if it doesn't satisfy the partial-index predicate */
if (pred != NULL) {
#ifndef OMIT_PARTIAL_INDEX
/* SetSlotContents(slot, htup); */
slot->val = htup;
if (ExecQual((List*)pred, econtext) == false)
continue;
#endif /* OMIT_PARTIAL_INDEX */
}
nitups++;
/*
* For the current heap tuple, extract all the attributes
* we use in this index, and note which are null.
*/
for (i = 1; i <= natts; i++) {
int attoff;
bool attnull;
/*
* Offsets are from the start of the tuple, and are
* zero-based; indices are one-based. The next call
* returns i - 1. That's data hiding for you.
*/
attoff = AttrNumberGetAttrOffset(i);
attdata[attoff] = GetIndexValue(htup,
htupdesc,
attoff,
attnum,
finfo,
&attnull,
buffer);
nulls[attoff] = (attnull ? 'n' : ' ');
}
/* form an index tuple and point it at the heap tuple */
itup = index_formtuple(itupdesc, attdata, nulls);
/*
* If the single index key is null, we don't insert it into
* the index. Btrees support scans on <, <=, =, >=, and >.
* Relational algebra says that A op B (where op is one of the
* operators above) returns null if either A or B is null. This
* means that no qualification used in an index scan could ever
* return true on a null attribute. It also means that indices
* can't be used by ISNULL or NOTNULL scans, but that's an
* artifact of the strategy map architecture chosen in 1986, not
* of the way nulls are handled here.
*/
/*
* New comments: NULLs handling.
* While we can't do NULL comparison, we can follow simple
* rule for ordering items on btree pages - NULLs greater
* NOT_NULLs and NULL = NULL is TRUE. Sure, it's just rule
* for placing/finding items and no more - keytest'll return
* FALSE for a = 5 for items having 'a' isNULL.
* Look at _bt_skeycmp, _bt_compare and _bt_itemcmp for
* how it works. - vadim 03/23/97
if (itup->t_info & INDEX_NULL_MASK) {
pfree(itup);
continue;
}
*/
itup->t_tid = htup->t_ctid;
btitem = _bt_formitem(itup);
/*
* if we are doing bottom-up btree build, we insert the index
* into a spool page for subsequent processing. otherwise, we
* insert into the btree.
*/
What looks like some *major* improvements to btree indexing... Patches from: aoki@CS.Berkeley.EDU (Paul M. Aoki) i gave jolly my btree bulkload code a long, long time ago but never gave him a bunch of my bugfixes. here's a diff against the 6.0 baseline. for some reason, this code has slowed down somewhat relative to the insertion-build code on very small tables. don't know why -- it used to be within about 10%. anyway, here are some (highly unscientific!) timings on a dec 3000/300 for synthetic tables with 10k, 100k and 1000k tuples (basically, 1mb, 10mb and 100mb heaps). 'c' means clustered (pre-sorted) inputs and 'u' means unclustered (randomly ordered) inputs. the 10k table basically fits in the buffer pool, but the 100k and 1000k tables don't. as you can see, insertion build is fine if you've sorted your heaps on your index key or if your heap fits in core, but is absolutely horrible on unordered data (yes, that's 7.5 hours to index 100mb of data...) because of the zillions of random i/os. if it doesn't work for you for whatever reason, you can always turn it back off by flipping the FastBuild flag in nbtree.c. i don't have time to maintain it. good luck! baseline code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 8.6 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 9.1 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.2 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 652.4 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.1 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 26772.9 bulkloading code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 11.3 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 10.4 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.5 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 63.5 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.9 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 701.0
1997-02-12 06:04:52 +01:00
if (usefast) {
_bt_spool(index, btitem, spool);
} else {
res = _bt_doinsert(index, btitem, isunique, heap);
}
pfree(btitem);
pfree(itup);
if (res) {
pfree(res);
}
}
/* okay, all heap tuples are indexed */
heap_endscan(hscan);
if (pred != NULL || oldPred != NULL) {
#ifndef OMIT_PARTIAL_INDEX
ExecDestroyTupleTable(tupleTable, true);
pfree(econtext);
#endif /* OMIT_PARTIAL_INDEX */
}
/*
* if we are doing bottom-up btree build, we now have a bunch of
* sorted runs in the spool pages. finish the build by (1)
* merging the runs, (2) inserting the sorted tuples into btree
* pages and (3) building the upper levels.
*/
What looks like some *major* improvements to btree indexing... Patches from: aoki@CS.Berkeley.EDU (Paul M. Aoki) i gave jolly my btree bulkload code a long, long time ago but never gave him a bunch of my bugfixes. here's a diff against the 6.0 baseline. for some reason, this code has slowed down somewhat relative to the insertion-build code on very small tables. don't know why -- it used to be within about 10%. anyway, here are some (highly unscientific!) timings on a dec 3000/300 for synthetic tables with 10k, 100k and 1000k tuples (basically, 1mb, 10mb and 100mb heaps). 'c' means clustered (pre-sorted) inputs and 'u' means unclustered (randomly ordered) inputs. the 10k table basically fits in the buffer pool, but the 100k and 1000k tables don't. as you can see, insertion build is fine if you've sorted your heaps on your index key or if your heap fits in core, but is absolutely horrible on unordered data (yes, that's 7.5 hours to index 100mb of data...) because of the zillions of random i/os. if it doesn't work for you for whatever reason, you can always turn it back off by flipping the FastBuild flag in nbtree.c. i don't have time to maintain it. good luck! baseline code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 8.6 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 9.1 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.2 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 652.4 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.1 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 26772.9 bulkloading code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 11.3 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 10.4 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.5 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 63.5 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.9 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 701.0
1997-02-12 06:04:52 +01:00
if (usefast) {
_bt_spool(index, (BTItem) NULL, spool); /* flush the spool */
_bt_leafbuild(index, spool);
_bt_spooldestroy(spool);
}
What looks like some *major* improvements to btree indexing... Patches from: aoki@CS.Berkeley.EDU (Paul M. Aoki) i gave jolly my btree bulkload code a long, long time ago but never gave him a bunch of my bugfixes. here's a diff against the 6.0 baseline. for some reason, this code has slowed down somewhat relative to the insertion-build code on very small tables. don't know why -- it used to be within about 10%. anyway, here are some (highly unscientific!) timings on a dec 3000/300 for synthetic tables with 10k, 100k and 1000k tuples (basically, 1mb, 10mb and 100mb heaps). 'c' means clustered (pre-sorted) inputs and 'u' means unclustered (randomly ordered) inputs. the 10k table basically fits in the buffer pool, but the 100k and 1000k tables don't. as you can see, insertion build is fine if you've sorted your heaps on your index key or if your heap fits in core, but is absolutely horrible on unordered data (yes, that's 7.5 hours to index 100mb of data...) because of the zillions of random i/os. if it doesn't work for you for whatever reason, you can always turn it back off by flipping the FastBuild flag in nbtree.c. i don't have time to maintain it. good luck! baseline code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 8.6 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 9.1 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.2 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 652.4 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.1 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 26772.9 bulkloading code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 11.3 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 10.4 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.5 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 63.5 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.9 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 701.0
1997-02-12 06:04:52 +01:00
#if 0
{
extern int ReadBufferCount, BufferHitCount, BufferFlushCount;
extern long NDirectFileRead, NDirectFileWrite;
printf("buffer(%d): r=%d w=%d\n", heap->rd_rel->relblocksz,
ReadBufferCount - BufferHitCount, BufferFlushCount);
printf("direct(%d): r=%d w=%d\n", LocalBlockSize,
NDirectFileRead, NDirectFileWrite);
}
#endif
/*
* Since we just counted the tuples in the heap, we update its
* stats in pg_class to guarantee that the planner takes advantage
* of the index we just created. Finally, only update statistics
* during normal index definitions, not for indices on system catalogs
* created during bootstrap processing. We must close the relations
* before updatings statistics to guarantee that the relcache entries
* are flushed when we increment the command counter in UpdateStats().
*/
if (IsNormalProcessingMode())
{
hrelid = heap->rd_id;
irelid = index->rd_id;
heap_close(heap);
index_close(index);
UpdateStats(hrelid, nhtups, true);
UpdateStats(irelid, nitups, false);
if (oldPred != NULL) {
if (nitups == nhtups) pred = NULL;
UpdateIndexPredicate(irelid, oldPred, pred);
}
}
pfree(nulls);
pfree(attdata);
/* all done */
BuildingBtree = false;
}
/*
* btinsert() -- insert an index tuple into a btree.
*
* Descend the tree recursively, find the appropriate location for our
* new tuple, put it there, set its unique OID as appropriate, and
* return an InsertIndexResult to the caller.
*/
InsertIndexResult
btinsert(Relation rel, Datum *datum, char *nulls, ItemPointer ht_ctid, Relation heapRel)
{
BTItem btitem;
IndexTuple itup;
InsertIndexResult res;
/* generate an index tuple */
itup = index_formtuple(RelationGetTupleDescriptor(rel), datum, nulls);
itup->t_tid = *ht_ctid;
/*
* See comments in btbuild.
if (itup->t_info & INDEX_NULL_MASK)
return ((InsertIndexResult) NULL);
*/
btitem = _bt_formitem(itup);
res = _bt_doinsert(rel, btitem,
IndexIsUnique(RelationGetRelationId(rel)), heapRel);
pfree(btitem);
pfree(itup);
What looks like some *major* improvements to btree indexing... Patches from: aoki@CS.Berkeley.EDU (Paul M. Aoki) i gave jolly my btree bulkload code a long, long time ago but never gave him a bunch of my bugfixes. here's a diff against the 6.0 baseline. for some reason, this code has slowed down somewhat relative to the insertion-build code on very small tables. don't know why -- it used to be within about 10%. anyway, here are some (highly unscientific!) timings on a dec 3000/300 for synthetic tables with 10k, 100k and 1000k tuples (basically, 1mb, 10mb and 100mb heaps). 'c' means clustered (pre-sorted) inputs and 'u' means unclustered (randomly ordered) inputs. the 10k table basically fits in the buffer pool, but the 100k and 1000k tables don't. as you can see, insertion build is fine if you've sorted your heaps on your index key or if your heap fits in core, but is absolutely horrible on unordered data (yes, that's 7.5 hours to index 100mb of data...) because of the zillions of random i/os. if it doesn't work for you for whatever reason, you can always turn it back off by flipping the FastBuild flag in nbtree.c. i don't have time to maintain it. good luck! baseline code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 8.6 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 9.1 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.2 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 652.4 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.1 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 26772.9 bulkloading code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 11.3 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 10.4 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.5 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 63.5 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.9 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 701.0
1997-02-12 06:04:52 +01:00
/* adjust any active scans that will be affected by this insertion */
_bt_adjscans(rel, &(res->pointerData), BT_INSERT);
return (res);
}
/*
* btgettuple() -- Get the next tuple in the scan.
*/
char *
btgettuple(IndexScanDesc scan, ScanDirection dir)
{
RetrieveIndexResult res;
/*
* If we've already initialized this scan, we can just advance it
* in the appropriate direction. If we haven't done so yet, we
* call a routine to get the first item in the scan.
*/
if (ItemPointerIsValid(&(scan->currentItemData)))
res = _bt_next(scan, dir);
else
res = _bt_first(scan, dir);
return ((char *) res);
}
/*
* btbeginscan() -- start a scan on a btree index
*/
char *
btbeginscan(Relation rel, bool fromEnd, uint16 keysz, ScanKey scankey)
{
IndexScanDesc scan;
/* get the scan */
scan = RelationGetIndexScan(rel, fromEnd, keysz, scankey);
/* register scan in case we change pages it's using */
_bt_regscan(scan);
return ((char *) scan);
}
/*
* btrescan() -- rescan an index relation
*/
void
btrescan(IndexScanDesc scan, bool fromEnd, ScanKey scankey)
{
ItemPointer iptr;
BTScanOpaque so;
StrategyNumber strat;
so = (BTScanOpaque) scan->opaque;
/* we hold a read lock on the current page in the scan */
if (ItemPointerIsValid(iptr = &(scan->currentItemData))) {
_bt_relbuf(scan->relation, so->btso_curbuf, BT_READ);
so->btso_curbuf = InvalidBuffer;
ItemPointerSetInvalid(iptr);
}
/* and we hold a read lock on the last marked item in the scan */
if (ItemPointerIsValid(iptr = &(scan->currentMarkData))) {
_bt_relbuf(scan->relation, so->btso_mrkbuf, BT_READ);
so->btso_mrkbuf = InvalidBuffer;
ItemPointerSetInvalid(iptr);
}
if ( so == NULL ) /* if called from btbeginscan */
{
so = (BTScanOpaque) palloc(sizeof(BTScanOpaqueData));
so->btso_curbuf = so->btso_mrkbuf = InvalidBuffer;
so->keyData = (ScanKey) NULL;
if ( scan->numberOfKeys > 0)
so->keyData = (ScanKey) palloc (scan->numberOfKeys * sizeof(ScanKeyData));
scan->opaque = so;
scan->flags = 0x0;
}
/* reset the scan key */
so->numberOfKeys = scan->numberOfKeys;
so->numberOfFirstKeys = 0; /* may be changed by _bt_orderkeys */
so->qual_ok = 1; /* may be changed by _bt_orderkeys */
if (scan->numberOfKeys > 0) {
memmove(scan->keyData,
scankey,
scan->numberOfKeys * sizeof(ScanKeyData));
memmove(so->keyData,
scankey,
so->numberOfKeys * sizeof(ScanKeyData));
/* order the keys in the qualification */
_bt_orderkeys(scan->relation, so);
}
/* finally, be sure that the scan exploits the tree order */
scan->scanFromEnd = false;
if ( so->numberOfKeys > 0 ) {
strat = _bt_getstrat(scan->relation, 1 /* XXX */,
so->keyData[0].sk_procedure);
if (strat == BTLessStrategyNumber
|| strat == BTLessEqualStrategyNumber)
scan->scanFromEnd = true;
} else {
scan->scanFromEnd = true;
}
}
void
btmovescan(IndexScanDesc scan, Datum v)
{
ItemPointer iptr;
BTScanOpaque so;
so = (BTScanOpaque) scan->opaque;
/* release any locks we still hold */
if (ItemPointerIsValid(iptr = &(scan->currentItemData))) {
_bt_relbuf(scan->relation, so->btso_curbuf, BT_READ);
so->btso_curbuf = InvalidBuffer;
ItemPointerSetInvalid(iptr);
}
/* scan->keyData[0].sk_argument = v; */
so->keyData[0].sk_argument = v;
}
/*
* btendscan() -- close down a scan
*/
void
btendscan(IndexScanDesc scan)
{
ItemPointer iptr;
BTScanOpaque so;
so = (BTScanOpaque) scan->opaque;
/* release any locks we still hold */
if (ItemPointerIsValid(iptr = &(scan->currentItemData))) {
if (BufferIsValid(so->btso_curbuf))
_bt_relbuf(scan->relation, so->btso_curbuf, BT_READ);
so->btso_curbuf = InvalidBuffer;
ItemPointerSetInvalid(iptr);
}
if (ItemPointerIsValid(iptr = &(scan->currentMarkData))) {
if (BufferIsValid(so->btso_mrkbuf))
_bt_relbuf(scan->relation, so->btso_mrkbuf, BT_READ);
so->btso_mrkbuf = InvalidBuffer;
ItemPointerSetInvalid(iptr);
}
1996-11-15 19:37:10 +01:00
if ( so->keyData != (ScanKey) NULL )
pfree (so->keyData);
pfree (so);
1996-11-15 19:37:10 +01:00
_bt_dropscan(scan);
}
/*
* btmarkpos() -- save current scan position
*/
void
btmarkpos(IndexScanDesc scan)
{
ItemPointer iptr;
BTScanOpaque so;
so = (BTScanOpaque) scan->opaque;
/* release lock on old marked data, if any */
if (ItemPointerIsValid(iptr = &(scan->currentMarkData))) {
_bt_relbuf(scan->relation, so->btso_mrkbuf, BT_READ);
so->btso_mrkbuf = InvalidBuffer;
ItemPointerSetInvalid(iptr);
}
/* bump lock on currentItemData and copy to currentMarkData */
if (ItemPointerIsValid(&(scan->currentItemData))) {
so->btso_mrkbuf = _bt_getbuf(scan->relation,
BufferGetBlockNumber(so->btso_curbuf),
BT_READ);
scan->currentMarkData = scan->currentItemData;
}
}
/*
* btrestrpos() -- restore scan to last saved position
*/
void
btrestrpos(IndexScanDesc scan)
{
ItemPointer iptr;
BTScanOpaque so;
so = (BTScanOpaque) scan->opaque;
/* release lock on current data, if any */
if (ItemPointerIsValid(iptr = &(scan->currentItemData))) {
_bt_relbuf(scan->relation, so->btso_curbuf, BT_READ);
so->btso_curbuf = InvalidBuffer;
ItemPointerSetInvalid(iptr);
}
/* bump lock on currentMarkData and copy to currentItemData */
if (ItemPointerIsValid(&(scan->currentMarkData))) {
so->btso_curbuf = _bt_getbuf(scan->relation,
BufferGetBlockNumber(so->btso_mrkbuf),
BT_READ);
scan->currentItemData = scan->currentMarkData;
}
}
/* stubs */
void
btdelete(Relation rel, ItemPointer tid)
{
/* adjust any active scans that will be affected by this deletion */
What looks like some *major* improvements to btree indexing... Patches from: aoki@CS.Berkeley.EDU (Paul M. Aoki) i gave jolly my btree bulkload code a long, long time ago but never gave him a bunch of my bugfixes. here's a diff against the 6.0 baseline. for some reason, this code has slowed down somewhat relative to the insertion-build code on very small tables. don't know why -- it used to be within about 10%. anyway, here are some (highly unscientific!) timings on a dec 3000/300 for synthetic tables with 10k, 100k and 1000k tuples (basically, 1mb, 10mb and 100mb heaps). 'c' means clustered (pre-sorted) inputs and 'u' means unclustered (randomly ordered) inputs. the 10k table basically fits in the buffer pool, but the 100k and 1000k tables don't. as you can see, insertion build is fine if you've sorted your heaps on your index key or if your heap fits in core, but is absolutely horrible on unordered data (yes, that's 7.5 hours to index 100mb of data...) because of the zillions of random i/os. if it doesn't work for you for whatever reason, you can always turn it back off by flipping the FastBuild flag in nbtree.c. i don't have time to maintain it. good luck! baseline code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 8.6 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 9.1 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.2 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 652.4 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.1 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 26772.9 bulkloading code: time psql -c 'create index c10 on k10 using btree (c int4_ops)' bttest real 11.3 time psql -c 'create index u10 on k10 using btree (b int4_ops)' bttest real 10.4 time psql -c 'create index c100 on k100 using btree (c int4_ops)' bttest real 59.5 time psql -c 'create index u100 on k100 using btree (b int4_ops)' bttest real 63.5 time psql -c 'create index c1000 on k1000 using btree (c int4_ops)' bttest real 636.9 time psql -c 'create index u1000 on k1000 using btree (b int4_ops)' bttest real 701.0
1997-02-12 06:04:52 +01:00
_bt_adjscans(rel, tid, BT_DELETE);
/* delete the data from the page */
_bt_pagedel(rel, tid);
}