Compare commits

..

5 Commits

Author SHA1 Message Date
Andrew Kane
915a630d00 Revert "Free tuple"
This reverts commit 466ec1ac46.
2023-10-22 16:09:07 -07:00
Andrew Kane
20c24e2d18 Added comment [skip ci] 2023-10-22 11:18:58 -07:00
Andrew Kane
466ec1ac46 Free tuple 2023-10-22 11:15:14 -07:00
Andrew Kane
ff2bdfb3b4 Improved check [skip ci] 2023-10-22 11:15:00 -07:00
Andrew Kane
0a3669a680 Added support for index-only scans to HNSW 2023-09-02 18:54:46 -07:00
20 changed files with 263 additions and 421 deletions

View File

@@ -1,8 +1,3 @@
## 0.5.1 (unreleased)
- Improved performance of index scans for IVFFlat after updates and deletes
- Fixed locking for index scans for HNSW
## 0.5.0 (2023-08-28)
- Added HNSW index type

View File

@@ -8,7 +8,7 @@ HEADERS = src/vector.h
TESTS = $(wildcard test/sql/*.sql)
REGRESS = $(patsubst test/sql/%.sql,%,$(TESTS))
REGRESS_OPTS = --inputdir=test --load-extension=$(EXTENSION)
REGRESS_OPTS = --inputdir=test --load-extension=vector
OPTFLAGS = -march=native

View File

@@ -5,7 +5,7 @@ OBJS = src\hnsw.obj src\hnswbuild.obj src\hnswinsert.obj src\hnswscan.obj src\hn
HEADERS = src\vector.h
REGRESS = btree cast copy functions input ivfflat_cosine ivfflat_ip ivfflat_l2 ivfflat_options ivfflat_unlogged
REGRESS_OPTS = --inputdir=test --load-extension=$(EXTENSION)
REGRESS_OPTS = --inputdir=test --load-extension=vector
# For /arch flags
# https://learn.microsoft.com/en-us/cpp/build/reference/arch-minimum-cpu-architecture

View File

@@ -595,18 +595,12 @@ pgvector is available on [these providers](https://github.com/pgvector/pgvector/
## Upgrading
Install the latest version. Then in each database you want to upgrade, run:
Install the latest version and run:
```sql
ALTER EXTENSION vector UPDATE;
```
You can check the version in the current database with:
```sql
SELECT extversion FROM pg_extension WHERE extname = 'vector';
```
## Upgrade Notes
### 0.4.0

View File

@@ -91,7 +91,7 @@ hnswcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
MemSet(&costs, 0, sizeof(costs));
index = index_open(path->indexinfo->indexoid, NoLock);
HnswGetMetaPageInfo(index, &m, NULL);
m = HnswGetM(index);
index_close(index, NoLock);
/* Approximate entry level */
@@ -155,6 +155,15 @@ hnswvalidate(Oid opclassoid)
return true;
}
/*
* Checks if index-only scan is supported
*/
static bool
hnswcanreturn(Relation indexRelation, int attno)
{
return attno == 1 && !OidIsValid(index_getprocid(indexRelation, 1, HNSW_NORM_PROC));
}
/*
* Define index handler
*
@@ -196,7 +205,7 @@ hnswhandler(PG_FUNCTION_ARGS)
amroutine->aminsert = hnswinsert;
amroutine->ambulkdelete = hnswbulkdelete;
amroutine->amvacuumcleanup = hnswvacuumcleanup;
amroutine->amcanreturn = NULL;
amroutine->amcanreturn = hnswcanreturn;
amroutine->amcostestimate = hnswcostestimate;
amroutine->amoptions = hnswoptions;
amroutine->amproperty = NULL; /* TODO AMPROP_DISTANCE_ORDERABLE */

View File

@@ -57,7 +57,7 @@
/* PROGRESS_CREATEIDX_SUBPHASE_INITIALIZE is 1 */
#define PROGRESS_HNSW_PHASE_LOAD 2
#define HNSW_ELEMENT_TUPLE_SIZE(_datum) MAXALIGN(offsetof(HnswElementTupleData, value) + VARSIZE_ANY(_datum))
#define HNSW_ELEMENT_TUPLE_SIZE(_dim) MAXALIGN(offsetof(HnswElementTupleData, vec) + VECTOR_SIZE(_dim))
#define HNSW_NEIGHBOR_TUPLE_SIZE(level, m) MAXALIGN(offsetof(HnswNeighborTupleData, indextids) + ((level) + 2) * (m) * sizeof(ItemPointerData))
#define HnswPageGetOpaque(page) ((HnswPageOpaque) PageGetSpecialPointer(page))
@@ -96,13 +96,12 @@ typedef struct HnswElementData
List *heaptids;
uint8 level;
uint8 deleted;
bool loaded;
HnswNeighborArray *neighbors;
BlockNumber blkno;
OffsetNumber offno;
OffsetNumber neighborOffno;
BlockNumber neighborPage;
Datum value;
Vector *vec;
} HnswElementData;
typedef HnswElementData * HnswElement;
@@ -201,7 +200,7 @@ typedef struct HnswElementTupleData
ItemPointerData heaptids[HNSW_HEAPTIDS];
ItemPointerData neighbortid;
uint16 unused2;
char value[FLEXIBLE_ARRAY_MEMBER];
Vector vec;
} HnswElementTupleData;
typedef HnswElementTupleData * HnswElementTuple;
@@ -219,6 +218,7 @@ typedef HnswNeighborTupleData * HnswNeighborTuple;
typedef struct HnswScanOpaqueData
{
bool first;
Buffer buf;
List *w;
MemoryContext tmpCtx;
@@ -259,16 +259,15 @@ typedef struct HnswVacuumState
/* Methods */
int HnswGetM(Relation index);
int HnswGetEfConstruction(Relation index);
FmgrInfo *HnswOptionalProcInfo(Relation index, uint16 procnum);
FmgrInfo *HnswOptionalProcInfo(Relation rel, uint16 procnum);
bool HnswNormValue(FmgrInfo *procinfo, Oid collation, Datum *value, Vector * result);
void HnswCommitBuffer(Buffer buf, GenericXLogState *state);
Buffer HnswNewBuffer(Relation index, ForkNumber forkNum);
void HnswInitPage(Buffer buf, Page page);
void HnswInitRegisterPage(Relation index, Buffer *buf, Page *page, GenericXLogState **state);
void HnswInit(void);
List *HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *procinfo, Oid collation, int m, bool inserting, HnswElement skipElement);
List *HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *procinfo, Oid collation, bool loadVec, HnswElement skipElement);
HnswElement HnswGetEntryPoint(Relation index);
void HnswGetMetaPageInfo(Relation index, int *m, HnswElement * entryPoint);
HnswElement HnswInitElement(ItemPointer tid, int m, double ml, int maxLevel);
void HnswFreeElement(HnswElement element);
HnswElement HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno);
@@ -285,7 +284,7 @@ void HnswLoadElementFromTuple(HnswElement element, HnswElementTuple etup, bool
void HnswLoadElement(HnswElement element, float *distance, Datum *q, Relation index, FmgrInfo *procinfo, Oid collation, bool loadVec);
void HnswSetElementTuple(HnswElementTuple etup, HnswElement element);
void HnswUpdateConnection(HnswElement element, HnswCandidate * hc, int m, int lc, int *updateIdx, Relation index, FmgrInfo *procinfo, Oid collation);
void HnswLoadNeighbors(HnswElement element, Relation index, int m);
void HnswLoadNeighbors(HnswElement element, Relation index);
/* Index access methods */
IndexBuildResult *hnswbuild(Relation heap, Relation index, IndexInfo *indexInfo);

View File

@@ -8,7 +8,6 @@
#include "lib/pairingheap.h"
#include "nodes/pg_list.h"
#include "storage/bufmgr.h"
#include "utils/datum.h"
#include "utils/memutils.h"
#if PG_VERSION_NUM >= 140000
@@ -107,6 +106,8 @@ CreateElementPages(HnswBuildState * buildstate)
{
Relation index = buildstate->index;
ForkNumber forkNum = buildstate->forkNum;
int dimensions = buildstate->dimensions;
Size etupSize;
Size maxSize;
HnswElementTuple etup;
HnswNeighborTuple ntup;
@@ -118,9 +119,10 @@ CreateElementPages(HnswBuildState * buildstate)
/* Calculate sizes */
maxSize = BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(HnswPageOpaqueData));
etupSize = HNSW_ELEMENT_TUPLE_SIZE(dimensions);
/* Allocate once */
etup = palloc0(maxSize);
etup = palloc0(etupSize);
ntup = palloc0(maxSize);
/* Prepare first page */
@@ -132,14 +134,12 @@ CreateElementPages(HnswBuildState * buildstate)
foreach(lc, buildstate->elements)
{
HnswElement element = lfirst(lc);
Size etupSize;
Size ntupSize;
Size combinedSize;
HnswSetElementTuple(etup, element);
/* Calculate sizes */
etupSize = HNSW_ELEMENT_TUPLE_SIZE(element->value);
ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(element->level, buildstate->m);
combinedSize = etupSize + ntupSize + sizeof(ItemIdData);
@@ -276,15 +276,18 @@ InsertTuple(Relation index, Datum *values, HnswElement element, HnswBuildState *
int m = buildstate->m;
/* Detoast once for all calls */
element->value = PointerGetDatum(PG_DETOAST_DATUM(values[0]));
Datum value = PointerGetDatum(PG_DETOAST_DATUM(values[0]));
/* Normalize if needed */
if (buildstate->normprocinfo != NULL)
{
if (!HnswNormValue(buildstate->normprocinfo, collation, &element->value, buildstate->normvec))
if (!HnswNormValue(buildstate->normprocinfo, collation, &value, buildstate->normvec))
return false;
}
/* Copy value to element so accessible outside of memory context */
memcpy(element->vec, DatumGetVector(value), VECTOR_SIZE(buildstate->dimensions));
/* Insert element in graph */
HnswInsertElement(element, entryPoint, NULL, procinfo, collation, m, efConstruction, false);
@@ -360,6 +363,7 @@ BuildCallback(Relation index, CALLBACK_ITEM_POINTER, Datum *values,
/* Allocate necessary memory outside of memory context */
element = HnswInitElement(tid, buildstate->m, buildstate->ml, buildstate->maxLevel);
element->vec = palloc(VECTOR_SIZE(buildstate->dimensions));
/* Use memory context since detoast can allocate */
oldCtx = MemoryContextSwitchTo(buildstate->tmpCtx);
@@ -367,8 +371,9 @@ BuildCallback(Relation index, CALLBACK_ITEM_POINTER, Datum *values,
/* Insert tuple */
inserted = InsertTuple(index, values, element, buildstate, &dup);
/* Switch memory context */
/* Reset memory context */
MemoryContextSwitchTo(oldCtx);
MemoryContextReset(buildstate->tmpCtx);
/* Add outside memory context */
if (dup != NULL)
@@ -376,16 +381,9 @@ BuildCallback(Relation index, CALLBACK_ITEM_POINTER, Datum *values,
/* Add to buildstate or free */
if (inserted)
{
element->value = datumCopy(element->value, false, -1);
element->loaded = true;
buildstate->elements = lappend(buildstate->elements, element);
}
else
HnswFreeElement(element);
/* Reset memory context */
MemoryContextReset(buildstate->tmpCtx);
}
/*

View File

@@ -123,6 +123,7 @@ WriteNewElementPages(Relation index, HnswElement e, int m, BlockNumber insertPag
Size minCombinedSize;
HnswElementTuple etup;
BlockNumber currentPage = insertPage;
int dimensions = e->vec->dim;
HnswNeighborTuple ntup;
Buffer nbuf;
Page npage;
@@ -131,7 +132,7 @@ WriteNewElementPages(Relation index, HnswElement e, int m, BlockNumber insertPag
BlockNumber newInsertPage = InvalidBlockNumber;
/* Calculate sizes */
etupSize = HNSW_ELEMENT_TUPLE_SIZE(e->value);
etupSize = HNSW_ELEMENT_TUPLE_SIZE(dimensions);
ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(e->level, m);
combinedSize = etupSize + ntupSize + sizeof(ItemIdData);
maxSize = BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(HnswPageOpaqueData));
@@ -328,7 +329,7 @@ HnswUpdateNeighborPages(Relation index, FmgrInfo *procinfo, Oid collation, HnswE
/* Get latest neighbors since they may have changed */
/* Do not lock yet since selecting neighbors can take time */
HnswLoadNeighbors(hc->element, index, m);
HnswLoadNeighbors(hc->element, index);
/*
* Could improve performance for vacuuming by checking neighbors
@@ -410,7 +411,7 @@ HnswAddDuplicate(Relation index, HnswElement element, HnswElement dup)
Buffer buf;
Page page;
GenericXLogState *state;
Size etupSize = HNSW_ELEMENT_TUPLE_SIZE(dup->value);
Size etupSize = HNSW_ELEMENT_TUPLE_SIZE(dup->vec->dim);
HnswElementTuple etup;
int i;
@@ -491,8 +492,9 @@ HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_ti
FmgrInfo *normprocinfo;
HnswElement entryPoint;
HnswElement element;
int m;
int m = HnswGetM(index);
int efConstruction = HnswGetEfConstruction(index);
double ml = HnswGetMl(m);
FmgrInfo *procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC);
Oid collation = index->rd_indcollation[0];
HnswElement dup;
@@ -509,6 +511,10 @@ HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_ti
return false;
}
/* Create an element */
element = HnswInitElement(heap_tid, m, ml, HnswGetMaxLevel(m));
element->vec = DatumGetVector(value);
/*
* Get a shared lock. This allows vacuum to ensure no in-flight inserts
* before repairing graph. Use a page lock so it does not interfere with
@@ -516,12 +522,8 @@ HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_ti
*/
LockPage(index, HNSW_UPDATE_LOCK, lockmode);
/* Get m and entry point */
HnswGetMetaPageInfo(index, &m, &entryPoint);
/* Create an element */
element = HnswInitElement(heap_tid, m, HnswGetMl(m), HnswGetMaxLevel(m));
element->value = value;
/* Get entry point */
entryPoint = HnswGetEntryPoint(index);
/* Prevent concurrent inserts when likely updating entry point */
if (entryPoint == NULL || element->level > entryPoint->level)

View File

@@ -17,26 +17,23 @@ GetScanItems(IndexScanDesc scan, Datum q)
Relation index = scan->indexRelation;
FmgrInfo *procinfo = so->procinfo;
Oid collation = so->collation;
bool loadVec = scan->xs_want_itup;
List *ep;
List *w;
int m;
HnswElement entryPoint;
/* Get m and entry point */
HnswGetMetaPageInfo(index, &m, &entryPoint);
HnswElement entryPoint = HnswGetEntryPoint(index);
if (entryPoint == NULL)
return NIL;
ep = list_make1(HnswEntryCandidate(entryPoint, q, index, procinfo, collation, false));
ep = list_make1(HnswEntryCandidate(entryPoint, q, index, procinfo, collation, loadVec));
for (int lc = entryPoint->level; lc >= 1; lc--)
{
w = HnswSearchLayer(q, ep, 1, lc, index, procinfo, collation, m, false, NULL);
w = HnswSearchLayer(q, ep, 1, lc, index, procinfo, collation, loadVec, NULL);
ep = w;
}
return HnswSearchLayer(q, ep, hnsw_ef_search, 0, index, procinfo, collation, m, false, NULL);
return HnswSearchLayer(q, ep, hnsw_ef_search, 0, index, procinfo, collation, loadVec, NULL);
}
/*
@@ -62,33 +59,6 @@ GetDimensions(Relation index)
return dimensions;
}
/*
* Get scan value
*/
static Datum
GetScanValue(IndexScanDesc scan)
{
HnswScanOpaque so = (HnswScanOpaque) scan->opaque;
Datum value;
if (scan->orderByData->sk_flags & SK_ISNULL)
value = PointerGetDatum(InitVector(GetDimensions(scan->indexRelation)));
else
{
value = scan->orderByData->sk_argument;
/* Value should not be compressed or toasted */
Assert(!VARATT_IS_COMPRESSED(DatumGetPointer(value)));
Assert(!VARATT_IS_EXTENDED(DatumGetPointer(value)));
/* Fine if normalization fails */
if (so->normprocinfo != NULL)
HnswNormValue(so->normprocinfo, so->collation, &value, NULL);
}
return value;
}
/*
* Prepare for an index scan
*/
@@ -101,6 +71,7 @@ hnswbeginscan(Relation index, int nkeys, int norderbys)
scan = RelationGetIndexScan(index, nkeys, norderbys);
so = (HnswScanOpaque) palloc(sizeof(HnswScanOpaqueData));
so->buf = InvalidBuffer;
so->first = true;
so->tmpCtx = AllocSetContextCreate(CurrentMemoryContext,
"Hnsw scan temporary context",
@@ -113,11 +84,8 @@ hnswbeginscan(Relation index, int nkeys, int norderbys)
scan->opaque = so;
/*
* Get a shared lock. This allows vacuum to ensure no in-flight scans
* before marking tuples as deleted.
*/
LockPage(scan->indexRelation, HNSW_SCAN_LOCK, ShareLock);
/* OK to always set since cheap */
scan->xs_itupdesc = RelationGetDescr(index);
return scan;
}
@@ -166,18 +134,40 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
if (scan->orderByData == NULL)
elog(ERROR, "cannot scan hnsw index without order");
/* Get scan value */
value = GetScanValue(scan);
if (scan->orderByData->sk_flags & SK_ISNULL)
value = PointerGetDatum(InitVector(GetDimensions(scan->indexRelation)));
else
{
value = scan->orderByData->sk_argument;
/* Value should not be compressed or toasted */
Assert(!VARATT_IS_COMPRESSED(DatumGetPointer(value)));
Assert(!VARATT_IS_EXTENDED(DatumGetPointer(value)));
/* Fine if normalization fails */
if (so->normprocinfo != NULL)
HnswNormValue(so->normprocinfo, so->collation, &value, NULL);
}
/*
* Get a shared lock. This allows vacuum to ensure no in-flight scans
* before marking tuples as deleted.
*/
LockPage(scan->indexRelation, HNSW_SCAN_LOCK, ShareLock);
so->w = GetScanItems(scan, value);
/* Release shared lock */
UnlockPage(scan->indexRelation, HNSW_SCAN_LOCK, ShareLock);
so->first = false;
}
while (list_length(so->w) > 0)
{
HnswCandidate *hc = llast(so->w);
ItemPointer heaptid;
ItemPointer tid;
BlockNumber indexblkno;
/* Move to next element if no valid heap TIDs */
if (list_length(hc->element->heaptids) == 0)
@@ -186,26 +176,39 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
continue;
}
heaptid = llast(hc->element->heaptids);
tid = llast(hc->element->heaptids);
indexblkno = hc->element->blkno;
hc->element->heaptids = list_delete_last(hc->element->heaptids);
if (scan->xs_want_itup)
{
Datum value = PointerGetDatum(hc->element->vec);
bool isnull = false;
scan->xs_itup = index_form_tuple(scan->xs_itupdesc, &value, &isnull);
scan->xs_itup->t_tid = *tid;
}
MemoryContextSwitchTo(oldCtx);
#if PG_VERSION_NUM >= 120000
scan->xs_heaptid = *heaptid;
scan->xs_heaptid = *tid;
#else
scan->xs_ctup.t_self = *heaptid;
scan->xs_ctup.t_self = *tid;
#endif
/* Unpin buffer */
if (BufferIsValid(so->buf))
ReleaseBuffer(so->buf);
/*
* Typically, an index scan must maintain a pin on the index page
* holding the item last returned by amgettuple. However, this is not
* needed with the current vacuum strategy, which ensures scans do not
* visit tuples in danger of being marked as deleted.
* An index scan must maintain a pin on the index page holding the
* item last returned by amgettuple
*
* https://www.postgresql.org/docs/current/index-locking.html
*/
so->buf = ReadBuffer(scan->indexRelation, indexblkno);
scan->xs_recheckorderby = false;
return true;
@@ -223,8 +226,9 @@ hnswendscan(IndexScanDesc scan)
{
HnswScanOpaque so = (HnswScanOpaque) scan->opaque;
/* Release shared lock */
UnlockPage(scan->indexRelation, HNSW_SCAN_LOCK, ShareLock);
/* Release pin */
if (BufferIsValid(so->buf))
ReleaseBuffer(so->buf);
MemoryContextDelete(so->tmpCtx);

View File

@@ -4,7 +4,6 @@
#include "hnsw.h"
#include "storage/bufmgr.h"
#include "utils/datum.h"
#include "vector.h"
/*
@@ -39,12 +38,12 @@ HnswGetEfConstruction(Relation index)
* Get proc
*/
FmgrInfo *
HnswOptionalProcInfo(Relation index, uint16 procnum)
HnswOptionalProcInfo(Relation rel, uint16 procnum)
{
if (!OidIsValid(index_getprocid(index, 1, procnum)))
if (!OidIsValid(index_getprocid(rel, 1, procnum)))
return NULL;
return index_getprocinfo(index, 1, procnum);
return index_getprocinfo(rel, 1, procnum);
}
/*
@@ -144,17 +143,6 @@ HnswInitNeighbors(HnswElement element, int m)
}
}
/*
* Free neighbors
*/
static void
HnswFreeNeighbors(HnswElement element)
{
for (int lc = 0; lc <= element->level; lc++)
pfree(element->neighbors[lc].items);
pfree(element->neighbors);
}
/*
* Allocate an element
*/
@@ -186,10 +174,11 @@ HnswInitElement(ItemPointer heaptid, int m, double ml, int maxLevel)
void
HnswFreeElement(HnswElement element)
{
HnswFreeNeighbors(element);
list_free_deep(element->heaptids);
if (element->loaded)
pfree(DatumGetPointer(element->value));
for (int lc = 0; lc <= element->level; lc++)
pfree(element->neighbors[lc].items);
pfree(element->neighbors);
pfree(element->vec);
pfree(element);
}
@@ -216,48 +205,30 @@ HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno)
element->blkno = blkno;
element->offno = offno;
element->neighbors = NULL;
element->loaded = false;
element->vec = NULL;
return element;
}
/*
* Get the metapage info
*/
void
HnswGetMetaPageInfo(Relation index, int *m, HnswElement * entryPoint)
{
Buffer buf;
Page page;
HnswMetaPage metap;
buf = ReadBuffer(index, HNSW_METAPAGE_BLKNO);
LockBuffer(buf, BUFFER_LOCK_SHARE);
page = BufferGetPage(buf);
metap = HnswPageGetMeta(page);
if (m != NULL)
*m = metap->m;
if (entryPoint != NULL)
{
if (BlockNumberIsValid(metap->entryBlkno))
*entryPoint = HnswInitElementFromBlock(metap->entryBlkno, metap->entryOffno);
else
*entryPoint = NULL;
}
UnlockReleaseBuffer(buf);
}
/*
* Get the entry point
*/
HnswElement
HnswGetEntryPoint(Relation index)
{
HnswElement entryPoint;
Buffer buf;
Page page;
HnswMetaPage metap;
HnswElement entryPoint = NULL;
HnswGetMetaPageInfo(index, NULL, &entryPoint);
buf = ReadBuffer(index, HNSW_METAPAGE_BLKNO);
LockBuffer(buf, BUFFER_LOCK_SHARE);
page = BufferGetPage(buf);
metap = HnswPageGetMeta(page);
if (BlockNumberIsValid(metap->entryBlkno))
entryPoint = HnswInitElementFromBlock(metap->entryBlkno, metap->entryOffno);
UnlockReleaseBuffer(buf);
return entryPoint;
}
@@ -326,7 +297,7 @@ HnswSetElementTuple(HnswElementTuple etup, HnswElement element)
else
ItemPointerSetInvalid(&etup->heaptids[i]);
}
memcpy(&etup->value, DatumGetPointer(element->value), VARSIZE_ANY(element->value));
memcpy(&etup->vec, element->vec, VECTOR_SIZE(element->vec->dim));
}
/*
@@ -366,9 +337,10 @@ HnswSetNeighborTuple(HnswNeighborTuple ntup, HnswElement e, int m)
* Load neighbors from page
*/
static void
LoadNeighborsFromPage(HnswElement element, Relation index, Page page, int m)
LoadNeighborsFromPage(HnswElement element, Relation index, Page page)
{
HnswNeighborTuple ntup = (HnswNeighborTuple) PageGetItem(page, PageGetItemId(page, element->neighborOffno));
int m = HnswGetM(index);
int neighborCount = (element->level + 2) * m;
Assert(HnswIsNeighborTuple(ntup));
@@ -409,7 +381,7 @@ LoadNeighborsFromPage(HnswElement element, Relation index, Page page, int m)
* Load neighbors
*/
void
HnswLoadNeighbors(HnswElement element, Relation index, int m)
HnswLoadNeighbors(HnswElement element, Relation index)
{
Buffer buf;
Page page;
@@ -418,7 +390,7 @@ HnswLoadNeighbors(HnswElement element, Relation index, int m)
LockBuffer(buf, BUFFER_LOCK_SHARE);
page = BufferGetPage(buf);
LoadNeighborsFromPage(element, index, page, m);
LoadNeighborsFromPage(element, index, page);
UnlockReleaseBuffer(buf);
}
@@ -449,10 +421,8 @@ HnswLoadElementFromTuple(HnswElement element, HnswElementTuple etup, bool loadHe
if (loadVec)
{
Datum value = PointerGetDatum(&etup->value);
element->value = datumCopy(value, false, -1);
element->loaded = true;
element->vec = palloc(VECTOR_SIZE(etup->vec.dim));
memcpy(element->vec, &etup->vec, VECTOR_SIZE(etup->vec.dim));
}
}
@@ -480,7 +450,7 @@ HnswLoadElement(HnswElement element, float *distance, Datum *q, Relation index,
/* Calculate distance */
if (distance != NULL)
*distance = (float) DatumGetFloat8(FunctionCall2Coll(procinfo, collation, *q, PointerGetDatum(&etup->value)));
*distance = (float) DatumGetFloat8(FunctionCall2Coll(procinfo, collation, *q, PointerGetDatum(&etup->vec)));
UnlockReleaseBuffer(buf);
}
@@ -491,7 +461,7 @@ HnswLoadElement(HnswElement element, float *distance, Datum *q, Relation index,
static float
GetCandidateDistance(HnswCandidate * hc, Datum q, FmgrInfo *procinfo, Oid collation)
{
return DatumGetFloat8(FunctionCall2Coll(procinfo, collation, q, hc->element->value));
return DatumGetFloat8(FunctionCall2Coll(procinfo, collation, q, PointerGetDatum(hc->element->vec)));
}
/*
@@ -573,7 +543,7 @@ AddToVisited(HTAB *v, HnswCandidate * hc, Relation index, bool *found)
* Algorithm 2 from paper
*/
List *
HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *procinfo, Oid collation, int m, bool inserting, HnswElement skipElement)
HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *procinfo, Oid collation, bool loadVec, HnswElement skipElement)
{
ListCell *lc2;
@@ -628,7 +598,7 @@ HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *pro
break;
if (c->element->neighbors == NULL)
HnswLoadNeighbors(c->element, index, m);
HnswLoadNeighbors(c->element, index);
/* Get the neighborhood at layer lc */
neighborhood = &c->element->neighbors[lc];
@@ -649,7 +619,7 @@ HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *pro
if (index == NULL)
eDistance = GetCandidateDistance(e, q, procinfo, collation);
else
HnswLoadElement(e->element, &eDistance, &q, index, procinfo, collation, inserting);
HnswLoadElement(e->element, &eDistance, &q, index, procinfo, collation, loadVec);
Assert(!e->element->deleted);
@@ -726,7 +696,7 @@ HnswGetDistance(HnswElement a, HnswElement b, int lc, FmgrInfo *procinfo, Oid co
}
}
return DatumGetFloat8(FunctionCall2Coll(procinfo, collation, a->value, b->value));
return DatumGetFloat8(FunctionCall2Coll(procinfo, collation, PointerGetDatum(a->vec), PointerGetDatum(b->vec)));
}
/*
@@ -809,7 +779,7 @@ HnswFindDuplicate(HnswElement e)
HnswCandidate *neighbor = &neighbors->items[i];
/* Exit early since ordered by distance */
if (!datumIsEqual(e->value, neighbor->element->value, false, -1))
if (vector_cmp_internal(e->vec, neighbor->element->vec) != 0)
break;
/* Check for space */
@@ -884,13 +854,13 @@ HnswUpdateConnection(HnswElement element, HnswCandidate * hc, int m, int lc, int
/* Load elements on insert */
if (index != NULL)
{
Datum q = hc->element->value;
Datum q = PointerGetDatum(hc->element->vec);
for (int i = 0; i < currentNeighbors->length; i++)
{
HnswCandidate *hc3 = &currentNeighbors->items[i];
if (!hc3->element->loaded)
if (hc3->element->vec == NULL)
HnswLoadElement(hc3->element, &hc3->distance, &q, index, procinfo, collation, true);
else
hc3->distance = GetCandidateDistance(hc3, q, procinfo, collation);
@@ -972,7 +942,7 @@ HnswInsertElement(HnswElement element, HnswElement entryPoint, Relation index, F
List *w;
int level = element->level;
int entryLevel;
Datum q = element->value;
Datum q = PointerGetDatum(element->vec);
HnswElement skipElement = existing ? element : NULL;
/* No neighbors if no entry point */
@@ -986,7 +956,7 @@ HnswInsertElement(HnswElement element, HnswElement entryPoint, Relation index, F
/* 1st phase: greedy search to insert level */
for (int lc = entryLevel; lc >= level + 1; lc--)
{
w = HnswSearchLayer(q, ep, 1, lc, index, procinfo, collation, m, true, skipElement);
w = HnswSearchLayer(q, ep, 1, lc, index, procinfo, collation, true, skipElement);
ep = w;
}
@@ -1004,7 +974,7 @@ HnswInsertElement(HnswElement element, HnswElement entryPoint, Relation index, F
List *neighbors;
List *lw;
w = HnswSearchLayer(q, ep, efConstruction, lc, index, procinfo, collation, m, true, skipElement);
w = HnswSearchLayer(q, ep, efConstruction, lc, index, procinfo, collation, true, skipElement);
/* Elements being deleted or skipped can help with search */
/* but should be removed before selecting neighbors */

View File

@@ -93,7 +93,7 @@ RemoveHeapTids(HnswVacuumState * vacuumstate)
if (itemUpdated)
{
Size etupSize = HNSW_ELEMENT_TUPLE_SIZE(PointerGetDatum(&etup->value));
Size etupSize = HNSW_ELEMENT_TUPLE_SIZE(etup->vec.dim);
/* Mark rest as invalid */
for (int i = idx; i < HNSW_HEAPTIDS; i++)
@@ -330,10 +330,7 @@ RepairGraph(HnswVacuumState * vacuumstate)
BufferAccessStrategy bas = vacuumstate->bas;
BlockNumber blkno = HNSW_HEAD_BLKNO;
/*
* Wait for inserts to complete. Inserts before this point may have
* neighbors about to be deleted. Inserts after this point will not.
*/
/* Wait for inserts to complete */
LockPage(index, HNSW_UPDATE_LOCK, ExclusiveLock);
UnlockPage(index, HNSW_UPDATE_LOCK, ExclusiveLock);
@@ -446,11 +443,7 @@ MarkDeleted(HnswVacuumState * vacuumstate)
Relation index = vacuumstate->index;
BufferAccessStrategy bas = vacuumstate->bas;
/*
* Wait for index scans to complete. Scans before this point may contain
* tuples about to be deleted. Scans after this point will not, since the
* graph has been repaired.
*/
/* Wait for selects to complete */
LockPage(index, HNSW_SCAN_LOCK, ExclusiveLock);
UnlockPage(index, HNSW_SCAN_LOCK, ExclusiveLock);
@@ -485,7 +478,6 @@ MarkDeleted(HnswVacuumState * vacuumstate)
HnswNeighborTuple ntup;
Size etupSize;
Size ntupSize;
Datum value;
Buffer nbuf;
Page npage;
BlockNumber neighborPage;
@@ -509,11 +501,8 @@ MarkDeleted(HnswVacuumState * vacuumstate)
if (ItemPointerIsValid(&etup->heaptids[0]))
continue;
/* Get datum */
value = PointerGetDatum(&etup->value);
/* Calculate sizes */
etupSize = HNSW_ELEMENT_TUPLE_SIZE(value);
etupSize = HNSW_ELEMENT_TUPLE_SIZE(etup->vec.dim);
ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(etup->level, vacuumstate->m);
/* Get neighbor page */
@@ -536,7 +525,7 @@ MarkDeleted(HnswVacuumState * vacuumstate)
/* Overwrite element */
etup->deleted = 1;
MemSet(&etup->value, 0, VARSIZE_ANY(value));
MemSet(&etup->vec.x, 0, etup->vec.dim * sizeof(float));
/* Overwrite neighbors */
for (int i = 0; i < ntup->count; i++)
@@ -593,6 +582,7 @@ InitVacuumState(HnswVacuumState * vacuumstate, IndexVacuumInfo *info, IndexBulkD
vacuumstate->stats = stats;
vacuumstate->callback = callback;
vacuumstate->callback_state = callback_state;
vacuumstate->m = HnswGetM(index);
vacuumstate->efConstruction = HnswGetEfConstruction(index);
vacuumstate->bas = GetAccessStrategy(BAS_BULKREAD);
vacuumstate->procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC);
@@ -602,9 +592,6 @@ InitVacuumState(HnswVacuumState * vacuumstate, IndexVacuumInfo *info, IndexBulkD
"Hnsw vacuum temporary context",
ALLOCSET_DEFAULT_SIZES);
/* Get m from metapage */
HnswGetMetaPageInfo(index, &vacuumstate->m, NULL);
/* Create hash table */
hash_ctl.keysize = sizeof(ItemPointerData);
hash_ctl.entrysize = sizeof(ItemPointerData);

View File

@@ -506,30 +506,29 @@ CreateListPages(Relation index, VectorArray centers, int dimensions,
Buffer buf;
Page page;
GenericXLogState *state;
OffsetNumber offno;
Size itemsz;
IvfflatList list;
list = palloc0(BLCKSZ);
itemsz = MAXALIGN(IVFFLAT_LIST_SIZE(dimensions));
list = palloc(itemsz);
buf = IvfflatNewBuffer(index, forkNum);
IvfflatInitRegisterPage(index, &buf, &page, &state);
for (int i = 0; i < lists; i++)
{
OffsetNumber offno;
Datum center = PointerGetDatum(VectorArrayGet(centers, i));
Size listSize = MAXALIGN(IVFFLAT_LIST_SIZE(center));
/* Load list */
list->startPage = InvalidBlockNumber;
list->insertPage = InvalidBlockNumber;
memcpy(&list->center, DatumGetPointer(center), VARSIZE_ANY(center));
memcpy(&list->center, VectorArrayGet(centers, i), VECTOR_SIZE(dimensions));
/* Ensure free space */
if (PageGetFreeSpace(page) < listSize)
if (PageGetFreeSpace(page) < itemsz)
IvfflatAppendPage(index, &buf, &page, &state, forkNum);
/* Add the item */
offno = PageAddItem(page, (Item) list, listSize, InvalidOffsetNumber, false, false);
offno = PageAddItem(page, (Item) list, itemsz, InvalidOffsetNumber, false, false);
if (offno == InvalidOffsetNumber)
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));

View File

@@ -71,7 +71,7 @@ ivfflatcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
int lists;
double ratio;
double spc_seq_page_cost;
Relation index;
Relation indexRel;
#if PG_VERSION_NUM < 120000
List *qinfos;
#endif
@@ -89,9 +89,9 @@ ivfflatcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
MemSet(&costs, 0, sizeof(costs));
index = index_open(path->indexinfo->indexoid, NoLock);
IvfflatGetMetaPageInfo(index, &lists, NULL);
index_close(index, NoLock);
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;

View File

@@ -52,7 +52,7 @@
#define PROGRESS_IVFFLAT_PHASE_ASSIGN 3
#define PROGRESS_IVFFLAT_PHASE_LOAD 4
#define IVFFLAT_LIST_SIZE(_datum) (offsetof(IvfflatListData, center) + VARSIZE_ANY(_datum))
#define IVFFLAT_LIST_SIZE(_dim) (offsetof(IvfflatListData, center) + VECTOR_SIZE(_dim))
#define IvfflatPageGetOpaque(page) ((IvfflatPageOpaque) PageGetSpecialPointer(page))
#define IvfflatPageGetMeta(page) ((IvfflatMetaPageData *) PageGetContents(page))
@@ -229,7 +229,7 @@ typedef struct IvfflatListData
{
BlockNumber startPage;
BlockNumber insertPage;
char center[FLEXIBLE_ARRAY_MEMBER];
Vector center;
} IvfflatListData;
typedef IvfflatListData * IvfflatList;
@@ -244,10 +244,8 @@ typedef struct IvfflatScanList
typedef struct IvfflatScanOpaqueData
{
int probes;
int dimensions;
bool first;
Buffer buf;
ItemPointerData heaptid;
/* Sorting */
Tuplesortstate *sortstate;
@@ -277,10 +275,9 @@ VectorArray VectorArrayInit(int maxlen, int dimensions);
void VectorArrayFree(VectorArray arr);
void PrintVectorArray(char *msg, VectorArray arr);
void IvfflatKmeans(Relation index, VectorArray samples, VectorArray centers);
FmgrInfo *IvfflatOptionalProcInfo(Relation index, uint16 procnum);
FmgrInfo *IvfflatOptionalProcInfo(Relation rel, uint16 procnum);
bool IvfflatNormValue(FmgrInfo *procinfo, Oid collation, Datum *value, Vector * result);
int IvfflatGetLists(Relation index);
void IvfflatGetMetaPageInfo(Relation index, int *lists, int *dimensions);
void IvfflatUpdateList(Relation index, ListInfo listInfo, BlockNumber insertPage, BlockNumber originalInsertPage, BlockNumber startPage, ForkNumber forkNum);
void IvfflatCommitBuffer(Buffer buf, GenericXLogState *state);
void IvfflatAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **state, ForkNumber forkNum);

View File

@@ -11,37 +11,36 @@
* Find the list that minimizes the distance function
*/
static void
FindInsertPage(Relation index, Datum *values, BlockNumber *insertPage, ListInfo * listInfo)
FindInsertPage(Relation rel, Datum *values, BlockNumber *insertPage, ListInfo * listInfo)
{
Buffer cbuf;
Page cpage;
IvfflatList list;
double distance;
double minDistance = DBL_MAX;
BlockNumber nextblkno = IVFFLAT_HEAD_BLKNO;
FmgrInfo *procinfo;
Oid collation;
OffsetNumber offno;
OffsetNumber maxoffno;
/* Avoid compiler warning */
listInfo->blkno = nextblkno;
listInfo->offno = FirstOffsetNumber;
procinfo = index_getprocinfo(index, 1, IVFFLAT_DISTANCE_PROC);
collation = index->rd_indcollation[0];
procinfo = index_getprocinfo(rel, 1, IVFFLAT_DISTANCE_PROC);
collation = rel->rd_indcollation[0];
/* Search all list pages */
while (BlockNumberIsValid(nextblkno))
{
Buffer cbuf;
Page cpage;
OffsetNumber maxoffno;
cbuf = ReadBuffer(index, nextblkno);
cbuf = ReadBuffer(rel, nextblkno);
LockBuffer(cbuf, BUFFER_LOCK_SHARE);
cpage = BufferGetPage(cbuf);
maxoffno = PageGetMaxOffsetNumber(cpage);
for (OffsetNumber offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
{
IvfflatList list;
double distance;
list = (IvfflatList) PageGetItem(cpage, PageGetItemId(cpage, offno));
distance = DatumGetFloat8(FunctionCall2Coll(procinfo, collation, values[0], PointerGetDatum(&list->center)));
@@ -64,7 +63,7 @@ FindInsertPage(Relation index, Datum *values, BlockNumber *insertPage, ListInfo
* Insert a tuple into the index
*/
static void
InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, Relation heapRel)
InsertTuple(Relation rel, Datum *values, bool *isnull, ItemPointer heap_tid, Relation heapRel)
{
IndexTuple itup;
Datum value;
@@ -81,20 +80,20 @@ InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, R
value = PointerGetDatum(PG_DETOAST_DATUM(values[0]));
/* Normalize if needed */
normprocinfo = IvfflatOptionalProcInfo(index, IVFFLAT_NORM_PROC);
normprocinfo = IvfflatOptionalProcInfo(rel, IVFFLAT_NORM_PROC);
if (normprocinfo != NULL)
{
if (!IvfflatNormValue(normprocinfo, index->rd_indcollation[0], &value, NULL))
if (!IvfflatNormValue(normprocinfo, rel->rd_indcollation[0], &value, NULL))
return;
}
/* Find the insert page - sets the page and list info */
FindInsertPage(index, values, &insertPage, &listInfo);
FindInsertPage(rel, values, &insertPage, &listInfo);
Assert(BlockNumberIsValid(insertPage));
originalInsertPage = insertPage;
/* Form tuple */
itup = index_form_tuple(RelationGetDescr(index), &value, isnull);
itup = index_form_tuple(RelationGetDescr(rel), &value, isnull);
itup->t_tid = *heap_tid;
/* Get tuple size */
@@ -104,10 +103,10 @@ InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, R
/* Find a page to insert the item */
for (;;)
{
buf = ReadBuffer(index, insertPage);
buf = ReadBuffer(rel, insertPage);
LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
state = GenericXLogStart(index);
state = GenericXLogStart(rel);
page = GenericXLogRegisterBuffer(state, buf, 0);
if (PageGetFreeSpace(page) >= itemsz)
@@ -127,9 +126,9 @@ InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, R
Page newpage;
/* Add a new page */
LockRelationForExtension(index, ExclusiveLock);
newbuf = IvfflatNewBuffer(index, MAIN_FORKNUM);
UnlockRelationForExtension(index, ExclusiveLock);
LockRelationForExtension(rel, ExclusiveLock);
newbuf = IvfflatNewBuffer(rel, MAIN_FORKNUM);
UnlockRelationForExtension(rel, ExclusiveLock);
/* Init new page */
newpage = GenericXLogRegisterBuffer(state, newbuf, GENERIC_XLOG_FULL_IMAGE);
@@ -150,7 +149,7 @@ InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, R
UnlockReleaseBuffer(buf);
/* Prepare new buffer */
state = GenericXLogStart(index);
state = GenericXLogStart(rel);
buf = newbuf;
page = GenericXLogRegisterBuffer(state, buf, 0);
break;
@@ -159,13 +158,13 @@ InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, R
/* Add to next offset */
if (PageAddItem(page, (Item) itup, itemsz, InvalidOffsetNumber, false, false) == InvalidOffsetNumber)
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(rel));
IvfflatCommitBuffer(buf, state);
/* Update the insert page */
if (insertPage != originalInsertPage)
IvfflatUpdateList(index, listInfo, insertPage, originalInsertPage, InvalidBlockNumber, MAIN_FORKNUM);
IvfflatUpdateList(rel, listInfo, insertPage, originalInsertPage, InvalidBlockNumber, MAIN_FORKNUM);
}
/*

View File

@@ -17,6 +17,10 @@ InitCenters(Relation index, VectorArray samples, VectorArray centers, float *low
FmgrInfo *procinfo;
Oid collation;
int64 j;
double distance;
double sum;
double choice;
Vector *vec;
float *weight = palloc(samples->length * sizeof(float));
int numCenters = centers->maxlen;
int numSamples = samples->length;
@@ -29,21 +33,17 @@ InitCenters(Relation index, VectorArray samples, VectorArray centers, float *low
centers->length++;
for (j = 0; j < numSamples; j++)
weight[j] = FLT_MAX;
weight[j] = DBL_MAX;
for (int i = 0; i < numCenters; i++)
{
double sum;
double choice;
CHECK_FOR_INTERRUPTS();
sum = 0.0;
for (j = 0; j < numSamples; j++)
{
Vector *vec = VectorArrayGet(samples, j);
double distance;
vec = VectorArrayGet(samples, j);
/* Only need to compute distance for new center */
/* TODO Use triangle inequality to reduce distance calculations */
@@ -112,6 +112,7 @@ CompareVectors(const void *a, const void *b)
static void
QuickCenters(Relation index, VectorArray samples, VectorArray centers)
{
Vector *vec;
int dimensions = centers->dim;
Oid collation = index->rd_indcollation[0];
FmgrInfo *normprocinfo = IvfflatOptionalProcInfo(index, IVFFLAT_KMEANS_NORM_PROC);
@@ -122,7 +123,7 @@ QuickCenters(Relation index, VectorArray samples, VectorArray centers)
qsort(samples->items, samples->length, VECTOR_SIZE(samples->dim), CompareVectors);
for (int i = 0; i < samples->length; i++)
{
Vector *vec = VectorArrayGet(samples, i);
vec = VectorArrayGet(samples, i);
if (i == 0 || CompareVectors(vec, VectorArrayGet(samples, i - 1)) != 0)
{
@@ -135,7 +136,7 @@ QuickCenters(Relation index, VectorArray samples, VectorArray centers)
/* Fill remaining with random data */
while (centers->length < centers->maxlen)
{
Vector *vec = VectorArrayGet(centers, centers->length);
vec = VectorArrayGet(centers, centers->length);
SET_VARSIZE(vec, VECTOR_SIZE(dimensions));
vec->dim = dimensions;
@@ -167,6 +168,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
Oid collation;
Vector *vec;
Vector *newCenter;
int iteration;
int64 j;
int64 k;
int dimensions = centers->dim;
@@ -180,6 +182,14 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
float *s;
float *halfcdist;
float *newcdist;
int changes;
double minDistance;
int closestCenter;
double distance;
bool rj;
bool rjreset;
double dxcx;
double dxc;
/* Calculate allocation sizes */
Size samplesSize = VECTOR_ARRAY_SIZE(samples->maxlen, samples->dim);
@@ -237,14 +247,14 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
/* Assign each x to its closest initial center c(x) = argmin d(x,c) */
for (j = 0; j < numSamples; j++)
{
float minDistance = FLT_MAX;
int closestCenter = 0;
minDistance = DBL_MAX;
closestCenter = 0;
/* Find closest center */
for (k = 0; k < numCenters; k++)
{
/* TODO Use Lemma 1 in k-means++ initialization */
float distance = lowerBound[j * numCenters + k];
distance = lowerBound[j * numCenters + k];
if (distance < minDistance)
{
@@ -258,14 +268,13 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
}
/* Give 500 iterations to converge */
for (int iteration = 0; iteration < 500; iteration++)
for (iteration = 0; iteration < 500; iteration++)
{
int changes = 0;
bool rjreset;
/* Can take a while, so ensure we can interrupt */
CHECK_FOR_INTERRUPTS();
changes = 0;
/* Step 1: For all centers, compute distance */
for (j = 0; j < numCenters; j++)
{
@@ -273,8 +282,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
for (k = j + 1; k < numCenters; k++)
{
float distance = 0.5 * DatumGetFloat8(FunctionCall2Coll(procinfo, collation, PointerGetDatum(vec), PointerGetDatum(VectorArrayGet(centers, k))));
distance = 0.5 * DatumGetFloat8(FunctionCall2Coll(procinfo, collation, PointerGetDatum(vec), PointerGetDatum(VectorArrayGet(centers, k))));
halfcdist[j * numCenters + k] = distance;
halfcdist[k * numCenters + j] = distance;
}
@@ -283,12 +291,10 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
/* For all centers c, compute s(c) */
for (j = 0; j < numCenters; j++)
{
float minDistance = FLT_MAX;
minDistance = DBL_MAX;
for (k = 0; k < numCenters; k++)
{
float distance;
if (j == k)
continue;
@@ -304,8 +310,6 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
for (j = 0; j < numSamples; j++)
{
bool rj;
/* Step 2: Identify all points x such that u(x) <= s(c(x)) */
if (upperBound[j] <= s[closestCenters[j]])
continue;
@@ -314,8 +318,6 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
for (k = 0; k < numCenters; k++)
{
float dxcx;
/* Step 3: For all remaining points x and centers c */
if (k == closestCenters[j])
continue;
@@ -345,7 +347,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
/* Step 3b */
if (dxcx > lowerBound[j * numCenters + k] || dxcx > halfcdist[closestCenters[j] * numCenters + k])
{
float dxc = DatumGetFloat8(FunctionCall2Coll(procinfo, collation, PointerGetDatum(vec), PointerGetDatum(VectorArrayGet(centers, k))));
dxc = DatumGetFloat8(FunctionCall2Coll(procinfo, collation, PointerGetDatum(vec), PointerGetDatum(VectorArrayGet(centers, k))));
/* d(x,c) calculated */
lowerBound[j * numCenters + k] = dxc;
@@ -359,6 +361,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
changes++;
}
}
}
}
@@ -375,8 +378,6 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
for (j = 0; j < numSamples; j++)
{
int closestCenter;
vec = VectorArrayGet(samples, j);
closestCenter = closestCenters[j];
@@ -425,7 +426,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
{
for (k = 0; k < numCenters; k++)
{
float distance = lowerBound[j * numCenters + k] - newcdist[k];
distance = lowerBound[j * numCenters + k] - newcdist[k];
if (distance < 0)
distance = 0;
@@ -441,7 +442,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
/* Step 7 */
for (j = 0; j < numCenters; j++)
VectorArraySet(centers, j, VectorArrayGet(newCenters, j));
memcpy(VectorArrayGet(centers, j), VectorArrayGet(newCenters, j), VECTOR_SIZE(dimensions));
if (changes == 0 && iteration != 0)
break;
@@ -464,6 +465,9 @@ static void
CheckCenters(Relation index, VectorArray centers)
{
FmgrInfo *normprocinfo;
Oid collation;
Vector *vec;
double norm;
if (centers->length != centers->maxlen)
elog(ERROR, "Not enough centers. Please report a bug.");
@@ -471,7 +475,7 @@ CheckCenters(Relation index, VectorArray centers)
/* Ensure no NaN or infinite values */
for (int i = 0; i < centers->length; i++)
{
Vector *vec = VectorArrayGet(centers, i);
vec = VectorArrayGet(centers, i);
for (int j = 0; j < vec->dim; j++)
{
@@ -497,12 +501,11 @@ CheckCenters(Relation index, VectorArray centers)
normprocinfo = IvfflatOptionalProcInfo(index, IVFFLAT_NORM_PROC);
if (normprocinfo != NULL)
{
Oid collation = index->rd_indcollation[0];
collation = index->rd_indcollation[0];
for (int i = 0; i < centers->length; i++)
{
double norm = DatumGetFloat8(FunctionCall1Coll(normprocinfo, collation, PointerGetDatum(VectorArrayGet(centers, i))));
norm = DatumGetFloat8(FunctionCall1Coll(normprocinfo, collation, PointerGetDatum(VectorArrayGet(centers, i))));
if (norm == 0)
elog(ERROR, "Zero norm detected. Please report a bug.");
}

View File

@@ -31,36 +31,36 @@ CompareLists(const pairingheap_node *a, const pairingheap_node *b, void *arg)
static void
GetScanLists(IndexScanDesc scan, Datum value)
{
IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
Buffer cbuf;
Page cpage;
IvfflatList list;
OffsetNumber offno;
OffsetNumber maxoffno;
BlockNumber nextblkno = IVFFLAT_HEAD_BLKNO;
int listCount = 0;
IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
double distance;
IvfflatScanList *scanlist;
double maxDistance = DBL_MAX;
/* Search all list pages */
while (BlockNumberIsValid(nextblkno))
{
Buffer cbuf;
Page cpage;
OffsetNumber maxoffno;
cbuf = ReadBuffer(scan->indexRelation, nextblkno);
LockBuffer(cbuf, BUFFER_LOCK_SHARE);
cpage = BufferGetPage(cbuf);
maxoffno = PageGetMaxOffsetNumber(cpage);
for (OffsetNumber offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
{
IvfflatList list = (IvfflatList) PageGetItem(cpage, PageGetItemId(cpage, offno));
double distance;
list = (IvfflatList) PageGetItem(cpage, PageGetItemId(cpage, offno));
/* Use procinfo from the index instead of scan key for performance */
distance = DatumGetFloat8(FunctionCall2Coll(so->procinfo, so->collation, PointerGetDatum(&list->center), value));
if (listCount < so->probes)
{
IvfflatScanList *scanlist;
scanlist = &so->lists[listCount];
scanlist->startPage = list->startPage;
scanlist->distance = distance;
@@ -75,8 +75,6 @@ GetScanLists(IndexScanDesc scan, Datum value)
}
else if (distance < maxDistance)
{
IvfflatScanList *scanlist;
/* Remove */
scanlist = (IvfflatScanList *) pairingheap_remove_first(so->listQueue);
@@ -103,6 +101,14 @@ static void
GetScanItems(IndexScanDesc scan, Datum value)
{
IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
Buffer buf;
Page page;
IndexTuple itup;
BlockNumber searchPage;
OffsetNumber offno;
OffsetNumber maxoffno;
Datum datum;
bool isnull;
TupleDesc tupdesc = RelationGetDescr(scan->indexRelation);
double tuples = 0;
@@ -122,32 +128,19 @@ GetScanItems(IndexScanDesc scan, Datum value)
/* Search closest probes lists */
while (!pairingheap_is_empty(so->listQueue))
{
BlockNumber searchPage = ((IvfflatScanList *) pairingheap_remove_first(so->listQueue))->startPage;
searchPage = ((IvfflatScanList *) pairingheap_remove_first(so->listQueue))->startPage;
/* Search all entry pages for list */
while (BlockNumberIsValid(searchPage))
{
Buffer buf;
Page page;
OffsetNumber maxoffno;
buf = ReadBufferExtended(scan->indexRelation, MAIN_FORKNUM, searchPage, RBM_NORMAL, bas);
LockBuffer(buf, BUFFER_LOCK_SHARE);
page = BufferGetPage(buf);
maxoffno = PageGetMaxOffsetNumber(page);
for (OffsetNumber offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
{
IndexTuple itup;
Datum datum;
bool isnull;
ItemId itemid = PageGetItemId(page, offno);
/* Skip dead tuples */
if (scan->ignore_killed_tuples && ItemIdIsDead(itemid))
continue;
itup = (IndexTuple) PageGetItem(page, itemid);
itup = (IndexTuple) PageGetItem(page, PageGetItemId(page, offno));
datum = index_getattr(itup, 1, tupdesc, &isnull);
/*
@@ -188,52 +181,26 @@ GetScanItems(IndexScanDesc scan, Datum value)
}
/*
* Mark prior tuple as dead
* Get dimensions from metapage
*/
static void
MarkPriorTupleDead(IndexScanDesc scan)
static int
GetDimensions(Relation index)
{
IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
Buffer buf = so->buf;
Buffer buf;
Page page;
OffsetNumber maxoffno;
IvfflatMetaPage metap;
int dimensions;
/* Safety check */
if (!BufferIsValid(so->buf) || !ItemPointerIsValid(&so->heaptid))
return;
/* Only a shared locked is needed for ItemIdMarkDead */
buf = ReadBuffer(index, IVFFLAT_METAPAGE_BLKNO);
LockBuffer(buf, BUFFER_LOCK_SHARE);
page = BufferGetPage(buf);
maxoffno = PageGetMaxOffsetNumber(page);
metap = IvfflatPageGetMeta(page);
for (OffsetNumber offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
{
ItemId itemid = PageGetItemId(page, offno);
IndexTuple itup = (IndexTuple) PageGetItem(page, itemid);
dimensions = metap->dimensions;
/*
* Find tuple. Since buffer has been pinned, tuple cannot have been
* vacuumed (and heap TID reused).
*/
if (ItemPointerEquals(&itup->t_tid, &so->heaptid))
{
/*
* Make sure tuple has not already been marked dead to avoid extra
* WAL if wal_log_hints or data checksums enabled
*/
if (!ItemIdIsDead(itemid))
{
ItemIdMarkDead(itemid);
MarkBufferDirtyHint(buf, true);
}
UnlockReleaseBuffer(buf);
break;
}
}
/* Unlock buffer */
LockBuffer(buf, BUFFER_LOCK_UNLOCK);
return dimensions;
}
/*
@@ -245,7 +212,6 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
IndexScanDesc scan;
IvfflatScanOpaque so;
int lists;
int dimensions;
AttrNumber attNums[] = {1};
Oid sortOperators[] = {Float8LessOperator};
Oid sortCollations[] = {InvalidOid};
@@ -253,9 +219,7 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
int probes = ivfflat_probes;
scan = RelationGetIndexScan(index, nkeys, norderbys);
/* Get lists and dimensions from metapage */
IvfflatGetMetaPageInfo(index, &lists, &dimensions);
lists = IvfflatGetLists(scan->indexRelation);
if (probes > lists)
probes = lists;
@@ -263,9 +227,7 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
so = (IvfflatScanOpaque) palloc(offsetof(IvfflatScanOpaqueData, lists) + probes * sizeof(IvfflatScanList));
so->buf = InvalidBuffer;
so->first = true;
ItemPointerSetInvalid(&so->heaptid);
so->probes = probes;
so->dimensions = dimensions;
/* Set support functions */
so->procinfo = index_getprocinfo(index, 1, IVFFLAT_DISTANCE_PROC);
@@ -279,7 +241,7 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
so->tupdesc = CreateTemplateTupleDesc(3, false);
#endif
TupleDescInitEntry(so->tupdesc, (AttrNumber) 1, "distance", FLOAT8OID, -1, 0);
TupleDescInitEntry(so->tupdesc, (AttrNumber) 2, "heaptid", TIDOID, -1, 0);
TupleDescInitEntry(so->tupdesc, (AttrNumber) 2, "tid", TIDOID, -1, 0);
TupleDescInitEntry(so->tupdesc, (AttrNumber) 3, "indexblkno", INT4OID, -1, 0);
/* Prep sort */
@@ -312,7 +274,6 @@ ivfflatrescan(IndexScanDesc scan, ScanKey keys, int nkeys, ScanKey orderbys, int
#endif
so->first = true;
ItemPointerSetInvalid(&so->heaptid);
pairingheap_reset(so->listQueue);
if (keys && scan->numberOfKeys > 0)
@@ -348,7 +309,7 @@ ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
elog(ERROR, "cannot scan ivfflat index without order");
if (scan->orderByData->sk_flags & SK_ISNULL)
value = PointerGetDatum(InitVector(so->dimensions));
value = PointerGetDatum(InitVector(GetDimensions(scan->indexRelation)));
else
{
value = scan->orderByData->sk_argument;
@@ -370,27 +331,18 @@ ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
if (value != scan->orderByData->sk_argument)
pfree(DatumGetPointer(value));
}
else
{
/* Mark prior tuple as dead */
if (scan->kill_prior_tuple)
MarkPriorTupleDead(scan);
}
if (tuplesort_gettupleslot(so->sortstate, true, false, so->slot, NULL))
{
ItemPointer heaptid = (ItemPointer) DatumGetPointer(slot_getattr(so->slot, 2, &so->isnull));
ItemPointer tid = (ItemPointer) DatumGetPointer(slot_getattr(so->slot, 2, &so->isnull));
BlockNumber indexblkno = DatumGetInt32(slot_getattr(so->slot, 3, &so->isnull));
#if PG_VERSION_NUM >= 120000
scan->xs_heaptid = *heaptid;
scan->xs_heaptid = *tid;
#else
scan->xs_ctup.t_self = *heaptid;
scan->xs_ctup.t_self = *tid;
#endif
/* Keep track of info needed to mark tuple as dead */
so->heaptid = *heaptid;
/* Unpin buffer */
if (BufferIsValid(so->buf))
ReleaseBuffer(so->buf);

View File

@@ -57,12 +57,12 @@ IvfflatGetLists(Relation index)
* Get proc
*/
FmgrInfo *
IvfflatOptionalProcInfo(Relation index, uint16 procnum)
IvfflatOptionalProcInfo(Relation rel, uint16 procnum)
{
if (!OidIsValid(index_getprocid(index, 1, procnum)))
if (!OidIsValid(index_getprocid(rel, 1, procnum)))
return NULL;
return index_getprocinfo(index, 1, procnum);
return index_getprocinfo(rel, 1, procnum);
}
/*
@@ -172,29 +172,6 @@ IvfflatAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **st
*buf = newbuf;
}
/*
* Get the metapage info
*/
void
IvfflatGetMetaPageInfo(Relation index, int *lists, int *dimensions)
{
Buffer buf;
Page page;
IvfflatMetaPage metap;
buf = ReadBuffer(index, IVFFLAT_METAPAGE_BLKNO);
LockBuffer(buf, BUFFER_LOCK_SHARE);
page = BufferGetPage(buf);
metap = IvfflatPageGetMeta(page);
*lists = metap->lists;
if (dimensions != NULL)
*dimensions = metap->dimensions;
UnlockReleaseBuffer(buf);
}
/*
* Update the start or insert page of a list
*/

View File

@@ -1,43 +0,0 @@
use strict;
use warnings;
use PostgresNode;
use TestLib;
use Test::More;
my $dim = 3;
my $array_sql = join(",", ('random()') x $dim);
# Initialize node
my $node = get_new_node('node');
$node->init;
$node->start;
# Create table and index
$node->safe_psql("postgres", "CREATE EXTENSION vector;");
$node->safe_psql("postgres", "CREATE TABLE tst (i serial, v vector($dim));");
$node->safe_psql("postgres",
"INSERT INTO tst (v) SELECT ARRAY[$array_sql] FROM generate_series(1, 10000) i;"
);
$node->safe_psql("postgres", "CREATE INDEX ON tst USING ivfflat (v vector_l2_ops);");
# Delete data
$node->safe_psql("postgres", "DELETE FROM tst WHERE i % 100 != 0;");
my $exp = $node->safe_psql("postgres", qq(
SET enable_indexscan = off;
SELECT i FROM tst ORDER BY v <-> '[0,0,0]';
));
# Run twice to make sure correct tuples marked as dead
for (1 .. 2)
{
my $res = $node->safe_psql("postgres", qq(
SET enable_seqscan = off;
SET ivfflat.probes = 100;
SELECT i FROM tst ORDER BY v <-> '[0,0,0]';
));
is($res, $exp);
}
done_testing();

View File

@@ -1,4 +1,4 @@
comment = 'vector data type and ivfflat and hnsw access methods'
comment = 'vector data type and ivfflat access method'
default_version = '0.5.0'
module_pathname = '$libdir/vector'
relocatable = true