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
24 changed files with 323 additions and 445 deletions

View File

@@ -8,8 +8,6 @@ jobs:
fail-fast: false fail-fast: false
matrix: matrix:
include: include:
- postgres: 17
os: ubuntu-22.04
- postgres: 16 - postgres: 16
os: ubuntu-22.04 os: ubuntu-22.04
- postgres: 15 - postgres: 15
@@ -23,7 +21,7 @@ jobs:
- postgres: 11 - postgres: 11
os: ubuntu-20.04 os: ubuntu-20.04
steps: steps:
- uses: actions/checkout@v4 - uses: actions/checkout@v3
- uses: ankane/setup-postgres@v1 - uses: ankane/setup-postgres@v1
with: with:
postgres-version: ${{ matrix.postgres }} postgres-version: ${{ matrix.postgres }}
@@ -45,7 +43,7 @@ jobs:
runs-on: macos-latest runs-on: macos-latest
if: ${{ !startsWith(github.ref_name, 'windows') }} if: ${{ !startsWith(github.ref_name, 'windows') }}
steps: steps:
- uses: actions/checkout@v4 - uses: actions/checkout@v3
- uses: ankane/setup-postgres@v1 - uses: ankane/setup-postgres@v1
with: with:
postgres-version: 14 postgres-version: 14
@@ -67,7 +65,7 @@ jobs:
runs-on: windows-latest runs-on: windows-latest
if: ${{ !startsWith(github.ref_name, 'mac') }} if: ${{ !startsWith(github.ref_name, 'mac') }}
steps: steps:
- uses: actions/checkout@v4 - uses: actions/checkout@v3
- uses: ankane/setup-postgres@v1 - uses: ankane/setup-postgres@v1
with: with:
postgres-version: 14 postgres-version: 14

View File

@@ -1,8 +1,3 @@
## 0.5.1 (2023-10-10)
- Improved performance of HNSW index builds
- Added check for MVCC-compliant snapshot for index scans
## 0.5.0 (2023-08-28) ## 0.5.0 (2023-08-28)
- Added HNSW index type - Added HNSW index type

View File

@@ -2,7 +2,7 @@
"name": "vector", "name": "vector",
"abstract": "Open-source vector similarity search for Postgres", "abstract": "Open-source vector similarity search for Postgres",
"description": "Supports L2 distance, inner product, and cosine distance", "description": "Supports L2 distance, inner product, and cosine distance",
"version": "0.5.1", "version": "0.5.0",
"maintainer": [ "maintainer": [
"Andrew Kane <andrew@ankane.org>" "Andrew Kane <andrew@ankane.org>"
], ],
@@ -20,7 +20,7 @@
"vector": { "vector": {
"file": "sql/vector.sql", "file": "sql/vector.sql",
"docfile": "README.md", "docfile": "README.md",
"version": "0.5.1", "version": "0.5.0",
"abstract": "Open-source vector similarity search for Postgres" "abstract": "Open-source vector similarity search for Postgres"
} }
}, },

View File

@@ -1,5 +1,5 @@
EXTENSION = vector EXTENSION = vector
EXTVERSION = 0.5.1 EXTVERSION = 0.5.0
MODULE_big = vector MODULE_big = vector
DATA = $(wildcard sql/*--*.sql) DATA = $(wildcard sql/*--*.sql)
@@ -8,7 +8,7 @@ HEADERS = src/vector.h
TESTS = $(wildcard test/sql/*.sql) TESTS = $(wildcard test/sql/*.sql)
REGRESS = $(patsubst test/sql/%.sql,%,$(TESTS)) REGRESS = $(patsubst test/sql/%.sql,%,$(TESTS))
REGRESS_OPTS = --inputdir=test --load-extension=$(EXTENSION) REGRESS_OPTS = --inputdir=test --load-extension=vector
OPTFLAGS = -march=native OPTFLAGS = -march=native

View File

@@ -1,11 +1,11 @@
EXTENSION = vector EXTENSION = vector
EXTVERSION = 0.5.1 EXTVERSION = 0.5.0
OBJS = src\hnsw.obj src\hnswbuild.obj src\hnswinsert.obj src\hnswscan.obj src\hnswutils.obj src\hnswvacuum.obj src\ivfbuild.obj src\ivfflat.obj src\ivfinsert.obj src\ivfkmeans.obj src\ivfscan.obj src\ivfutils.obj src\ivfvacuum.obj src\vector.obj OBJS = src\hnsw.obj src\hnswbuild.obj src\hnswinsert.obj src\hnswscan.obj src\hnswutils.obj src\hnswvacuum.obj src\ivfbuild.obj src\ivfflat.obj src\ivfinsert.obj src\ivfkmeans.obj src\ivfscan.obj src\ivfutils.obj src\ivfvacuum.obj src\vector.obj
HEADERS = src\vector.h HEADERS = src\vector.h
REGRESS = btree cast copy functions input ivfflat_cosine ivfflat_ip ivfflat_l2 ivfflat_options ivfflat_unlogged 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 # For /arch flags
# https://learn.microsoft.com/en-us/cpp/build/reference/arch-minimum-cpu-architecture # https://learn.microsoft.com/en-us/cpp/build/reference/arch-minimum-cpu-architecture

View File

@@ -18,7 +18,7 @@ Compile and install the extension (supports Postgres 11+)
```sh ```sh
cd /tmp cd /tmp
git clone --branch v0.5.1 https://github.com/pgvector/pgvector.git git clone --branch v0.5.0 https://github.com/pgvector/pgvector.git
cd pgvector cd pgvector
make make
make install # may need sudo make install # may need sudo
@@ -509,7 +509,7 @@ Then use `nmake` to build:
```cmd ```cmd
set "PGROOT=C:\Program Files\PostgreSQL\15" set "PGROOT=C:\Program Files\PostgreSQL\15"
git clone --branch v0.5.1 https://github.com/pgvector/pgvector.git git clone --branch v0.5.0 https://github.com/pgvector/pgvector.git
cd pgvector cd pgvector
nmake /F Makefile.win nmake /F Makefile.win
nmake /F Makefile.win install nmake /F Makefile.win install
@@ -530,7 +530,7 @@ This adds pgvector to the [Postgres image](https://hub.docker.com/_/postgres) (r
You can also build the image manually: You can also build the image manually:
```sh ```sh
git clone --branch v0.5.1 https://github.com/pgvector/pgvector.git git clone --branch v0.5.0 https://github.com/pgvector/pgvector.git
cd pgvector cd pgvector
docker build --build-arg PG_MAJOR=15 -t myuser/pgvector . docker build --build-arg PG_MAJOR=15 -t myuser/pgvector .
``` ```
@@ -595,18 +595,12 @@ pgvector is available on [these providers](https://github.com/pgvector/pgvector/
## Upgrading ## Upgrading
Install the latest version. Then in each database you want to upgrade, run: Install the latest version and run:
```sql ```sql
ALTER EXTENSION vector UPDATE; 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 ## Upgrade Notes
### 0.4.0 ### 0.4.0

View File

@@ -1,2 +0,0 @@
-- complain if script is sourced in psql, rather than via CREATE EXTENSION
\echo Use "ALTER EXTENSION vector UPDATE TO '0.5.1'" to load this file. \quit

View File

@@ -91,7 +91,7 @@ hnswcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
MemSet(&costs, 0, sizeof(costs)); MemSet(&costs, 0, sizeof(costs));
index = index_open(path->indexinfo->indexoid, NoLock); index = index_open(path->indexinfo->indexoid, NoLock);
HnswGetMetaPageInfo(index, &m, NULL); m = HnswGetM(index);
index_close(index, NoLock); index_close(index, NoLock);
/* Approximate entry level */ /* Approximate entry level */
@@ -155,6 +155,15 @@ hnswvalidate(Oid opclassoid)
return true; 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 * Define index handler
* *
@@ -196,7 +205,7 @@ hnswhandler(PG_FUNCTION_ARGS)
amroutine->aminsert = hnswinsert; amroutine->aminsert = hnswinsert;
amroutine->ambulkdelete = hnswbulkdelete; amroutine->ambulkdelete = hnswbulkdelete;
amroutine->amvacuumcleanup = hnswvacuumcleanup; amroutine->amvacuumcleanup = hnswvacuumcleanup;
amroutine->amcanreturn = NULL; amroutine->amcanreturn = hnswcanreturn;
amroutine->amcostestimate = hnswcostestimate; amroutine->amcostestimate = hnswcostestimate;
amroutine->amoptions = hnswoptions; amroutine->amoptions = hnswoptions;
amroutine->amproperty = NULL; /* TODO AMPROP_DISTANCE_ORDERABLE */ amroutine->amproperty = NULL; /* TODO AMPROP_DISTANCE_ORDERABLE */

View File

@@ -57,8 +57,6 @@
/* PROGRESS_CREATEIDX_SUBPHASE_INITIALIZE is 1 */ /* PROGRESS_CREATEIDX_SUBPHASE_INITIALIZE is 1 */
#define PROGRESS_HNSW_PHASE_LOAD 2 #define PROGRESS_HNSW_PHASE_LOAD 2
#define HNSW_MAX_SIZE (BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(HnswPageOpaqueData)) - sizeof(ItemIdData))
#define HNSW_ELEMENT_TUPLE_SIZE(_dim) MAXALIGN(offsetof(HnswElementTupleData, vec) + VECTOR_SIZE(_dim)) #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 HNSW_NEIGHBOR_TUPLE_SIZE(level, m) MAXALIGN(offsetof(HnswNeighborTupleData, indextids) + ((level) + 2) * (m) * sizeof(ItemPointerData))
@@ -112,13 +110,11 @@ typedef struct HnswCandidate
{ {
HnswElement element; HnswElement element;
float distance; float distance;
bool closer;
} HnswCandidate; } HnswCandidate;
typedef struct HnswNeighborArray typedef struct HnswNeighborArray
{ {
int length; int length;
bool closerSet;
HnswCandidate *items; HnswCandidate *items;
} HnswNeighborArray; } HnswNeighborArray;
@@ -222,6 +218,7 @@ typedef HnswNeighborTupleData * HnswNeighborTuple;
typedef struct HnswScanOpaqueData typedef struct HnswScanOpaqueData
{ {
bool first; bool first;
Buffer buf;
List *w; List *w;
MemoryContext tmpCtx; MemoryContext tmpCtx;
@@ -262,16 +259,15 @@ typedef struct HnswVacuumState
/* Methods */ /* Methods */
int HnswGetM(Relation index); int HnswGetM(Relation index);
int HnswGetEfConstruction(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); bool HnswNormValue(FmgrInfo *procinfo, Oid collation, Datum *value, Vector * result);
void HnswCommitBuffer(Buffer buf, GenericXLogState *state); void HnswCommitBuffer(Buffer buf, GenericXLogState *state);
Buffer HnswNewBuffer(Relation index, ForkNumber forkNum); Buffer HnswNewBuffer(Relation index, ForkNumber forkNum);
void HnswInitPage(Buffer buf, Page page); void HnswInitPage(Buffer buf, Page page);
void HnswInitRegisterPage(Relation index, Buffer *buf, Page *page, GenericXLogState **state); void HnswInitRegisterPage(Relation index, Buffer *buf, Page *page, GenericXLogState **state);
void HnswInit(void); 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); HnswElement HnswGetEntryPoint(Relation index);
void HnswGetMetaPageInfo(Relation index, int *m, HnswElement * entryPoint);
HnswElement HnswInitElement(ItemPointer tid, int m, double ml, int maxLevel); HnswElement HnswInitElement(ItemPointer tid, int m, double ml, int maxLevel);
void HnswFreeElement(HnswElement element); void HnswFreeElement(HnswElement element);
HnswElement HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno); HnswElement HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno);
@@ -288,7 +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 HnswLoadElement(HnswElement element, float *distance, Datum *q, Relation index, FmgrInfo *procinfo, Oid collation, bool loadVec);
void HnswSetElementTuple(HnswElementTuple etup, HnswElement element); 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 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 */ /* Index access methods */
IndexBuildResult *hnswbuild(Relation heap, Relation index, IndexInfo *indexInfo); IndexBuildResult *hnswbuild(Relation heap, Relation index, IndexInfo *indexInfo);

View File

@@ -81,6 +81,7 @@ HnswBuildAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **
HnswPageGetOpaque(*page)->nextblkno = BufferGetBlockNumber(newbuf); HnswPageGetOpaque(*page)->nextblkno = BufferGetBlockNumber(newbuf);
/* Commit */ /* Commit */
MarkBufferDirty(*buf);
GenericXLogFinish(*state); GenericXLogFinish(*state);
UnlockReleaseBuffer(*buf); UnlockReleaseBuffer(*buf);
@@ -117,12 +118,12 @@ CreateElementPages(HnswBuildState * buildstate)
ListCell *lc; ListCell *lc;
/* Calculate sizes */ /* Calculate sizes */
maxSize = HNSW_MAX_SIZE; maxSize = BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(HnswPageOpaqueData));
etupSize = HNSW_ELEMENT_TUPLE_SIZE(dimensions); etupSize = HNSW_ELEMENT_TUPLE_SIZE(dimensions);
/* Allocate once */ /* Allocate once */
etup = palloc0(etupSize); etup = palloc0(etupSize);
ntup = palloc0(BLCKSZ); ntup = palloc0(maxSize);
/* Prepare first page */ /* Prepare first page */
buf = HnswNewBuffer(index, forkNum); buf = HnswNewBuffer(index, forkNum);
@@ -178,6 +179,7 @@ CreateElementPages(HnswBuildState * buildstate)
insertPage = BufferGetBlockNumber(buf); insertPage = BufferGetBlockNumber(buf);
/* Commit */ /* Commit */
MarkBufferDirty(buf);
GenericXLogFinish(state); GenericXLogFinish(state);
UnlockReleaseBuffer(buf); UnlockReleaseBuffer(buf);
@@ -225,6 +227,7 @@ CreateNeighborPages(HnswBuildState * buildstate)
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index)); elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
/* Commit */ /* Commit */
MarkBufferDirty(buf);
GenericXLogFinish(state); GenericXLogFinish(state);
UnlockReleaseBuffer(buf); UnlockReleaseBuffer(buf);
} }

View File

@@ -135,7 +135,7 @@ WriteNewElementPages(Relation index, HnswElement e, int m, BlockNumber insertPag
etupSize = HNSW_ELEMENT_TUPLE_SIZE(dimensions); etupSize = HNSW_ELEMENT_TUPLE_SIZE(dimensions);
ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(e->level, m); ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(e->level, m);
combinedSize = etupSize + ntupSize + sizeof(ItemIdData); combinedSize = etupSize + ntupSize + sizeof(ItemIdData);
maxSize = HNSW_MAX_SIZE; maxSize = BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(HnswPageOpaqueData));
minCombinedSize = etupSize + HNSW_NEIGHBOR_TUPLE_SIZE(0, m) + sizeof(ItemIdData); minCombinedSize = etupSize + HNSW_NEIGHBOR_TUPLE_SIZE(0, m) + sizeof(ItemIdData);
/* Prepare element tuple */ /* Prepare element tuple */
@@ -202,6 +202,8 @@ WriteNewElementPages(Relation index, HnswElement e, int m, BlockNumber insertPag
HnswInsertAppendPage(index, &newbuf, &newpage, state, page); HnswInsertAppendPage(index, &newbuf, &newpage, state, page);
/* Commit */ /* Commit */
MarkBufferDirty(newbuf);
MarkBufferDirty(buf);
GenericXLogFinish(state); GenericXLogFinish(state);
/* Unlock previous buffer */ /* Unlock previous buffer */
@@ -268,6 +270,9 @@ WriteNewElementPages(Relation index, HnswElement e, int m, BlockNumber insertPag
} }
/* Commit */ /* Commit */
MarkBufferDirty(buf);
if (nbuf != buf)
MarkBufferDirty(nbuf);
GenericXLogFinish(state); GenericXLogFinish(state);
UnlockReleaseBuffer(buf); UnlockReleaseBuffer(buf);
if (nbuf != buf) if (nbuf != buf)
@@ -324,7 +329,7 @@ HnswUpdateNeighborPages(Relation index, FmgrInfo *procinfo, Oid collation, HnswE
/* Get latest neighbors since they may have changed */ /* Get latest neighbors since they may have changed */
/* Do not lock yet since selecting neighbors can take time */ /* 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 * Could improve performance for vacuuming by checking neighbors
@@ -386,6 +391,7 @@ HnswUpdateNeighborPages(Relation index, FmgrInfo *procinfo, Oid collation, HnswE
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index)); elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
/* Commit */ /* Commit */
MarkBufferDirty(buf);
GenericXLogFinish(state); GenericXLogFinish(state);
} }
else else
@@ -439,6 +445,7 @@ HnswAddDuplicate(Relation index, HnswElement element, HnswElement dup)
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index)); elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
/* Commit */ /* Commit */
MarkBufferDirty(buf);
GenericXLogFinish(state); GenericXLogFinish(state);
UnlockReleaseBuffer(buf); UnlockReleaseBuffer(buf);
@@ -485,8 +492,9 @@ HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_ti
FmgrInfo *normprocinfo; FmgrInfo *normprocinfo;
HnswElement entryPoint; HnswElement entryPoint;
HnswElement element; HnswElement element;
int m; int m = HnswGetM(index);
int efConstruction = HnswGetEfConstruction(index); int efConstruction = HnswGetEfConstruction(index);
double ml = HnswGetMl(m);
FmgrInfo *procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC); FmgrInfo *procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC);
Oid collation = index->rd_indcollation[0]; Oid collation = index->rd_indcollation[0];
HnswElement dup; HnswElement dup;
@@ -503,6 +511,10 @@ HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_ti
return false; 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 * 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 * before repairing graph. Use a page lock so it does not interfere with
@@ -510,12 +522,8 @@ HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_ti
*/ */
LockPage(index, HNSW_UPDATE_LOCK, lockmode); LockPage(index, HNSW_UPDATE_LOCK, lockmode);
/* Get m and entry point */ /* Get entry point */
HnswGetMetaPageInfo(index, &m, &entryPoint); entryPoint = HnswGetEntryPoint(index);
/* Create an element */
element = HnswInitElement(heap_tid, m, HnswGetMl(m), HnswGetMaxLevel(m));
element->vec = DatumGetVector(value);
/* Prevent concurrent inserts when likely updating entry point */ /* Prevent concurrent inserts when likely updating entry point */
if (entryPoint == NULL || element->level > entryPoint->level) if (entryPoint == NULL || element->level > entryPoint->level)

View File

@@ -17,26 +17,23 @@ GetScanItems(IndexScanDesc scan, Datum q)
Relation index = scan->indexRelation; Relation index = scan->indexRelation;
FmgrInfo *procinfo = so->procinfo; FmgrInfo *procinfo = so->procinfo;
Oid collation = so->collation; Oid collation = so->collation;
bool loadVec = scan->xs_want_itup;
List *ep; List *ep;
List *w; List *w;
int m; HnswElement entryPoint = HnswGetEntryPoint(index);
HnswElement entryPoint;
/* Get m and entry point */
HnswGetMetaPageInfo(index, &m, &entryPoint);
if (entryPoint == NULL) if (entryPoint == NULL)
return NIL; 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--) 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; 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; 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 * Prepare for an index scan
*/ */
@@ -101,6 +71,7 @@ hnswbeginscan(Relation index, int nkeys, int norderbys)
scan = RelationGetIndexScan(index, nkeys, norderbys); scan = RelationGetIndexScan(index, nkeys, norderbys);
so = (HnswScanOpaque) palloc(sizeof(HnswScanOpaqueData)); so = (HnswScanOpaque) palloc(sizeof(HnswScanOpaqueData));
so->buf = InvalidBuffer;
so->first = true; so->first = true;
so->tmpCtx = AllocSetContextCreate(CurrentMemoryContext, so->tmpCtx = AllocSetContextCreate(CurrentMemoryContext,
"Hnsw scan temporary context", "Hnsw scan temporary context",
@@ -113,6 +84,9 @@ hnswbeginscan(Relation index, int nkeys, int norderbys)
scan->opaque = so; scan->opaque = so;
/* OK to always set since cheap */
scan->xs_itupdesc = RelationGetDescr(index);
return scan; return scan;
} }
@@ -160,13 +134,20 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
if (scan->orderByData == NULL) if (scan->orderByData == NULL)
elog(ERROR, "cannot scan hnsw index without order"); elog(ERROR, "cannot scan hnsw index without order");
/* Requires MVCC-compliant snapshot as not able to maintain a pin */ if (scan->orderByData->sk_flags & SK_ISNULL)
/* https://www.postgresql.org/docs/current/index-locking.html */ value = PointerGetDatum(InitVector(GetDimensions(scan->indexRelation)));
if (!IsMVCCSnapshot(scan->xs_snapshot)) else
elog(ERROR, "non-MVCC snapshots are not supported with hnsw"); {
value = scan->orderByData->sk_argument;
/* Get scan value */ /* Value should not be compressed or toasted */
value = GetScanValue(scan); 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 * Get a shared lock. This allows vacuum to ensure no in-flight scans
@@ -185,7 +166,8 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
while (list_length(so->w) > 0) while (list_length(so->w) > 0)
{ {
HnswCandidate *hc = llast(so->w); HnswCandidate *hc = llast(so->w);
ItemPointer heaptid; ItemPointer tid;
BlockNumber indexblkno;
/* Move to next element if no valid heap TIDs */ /* Move to next element if no valid heap TIDs */
if (list_length(hc->element->heaptids) == 0) if (list_length(hc->element->heaptids) == 0)
@@ -194,18 +176,40 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
continue; continue;
} }
heaptid = llast(hc->element->heaptids); tid = llast(hc->element->heaptids);
indexblkno = hc->element->blkno;
hc->element->heaptids = list_delete_last(hc->element->heaptids); 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); MemoryContextSwitchTo(oldCtx);
#if PG_VERSION_NUM >= 120000 #if PG_VERSION_NUM >= 120000
scan->xs_heaptid = *heaptid; scan->xs_heaptid = *tid;
#else #else
scan->xs_ctup.t_self = *heaptid; scan->xs_ctup.t_self = *tid;
#endif #endif
/* Unpin buffer */
if (BufferIsValid(so->buf))
ReleaseBuffer(so->buf);
/*
* 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; scan->xs_recheckorderby = false;
return true; return true;
} }
@@ -222,6 +226,10 @@ hnswendscan(IndexScanDesc scan)
{ {
HnswScanOpaque so = (HnswScanOpaque) scan->opaque; HnswScanOpaque so = (HnswScanOpaque) scan->opaque;
/* Release pin */
if (BufferIsValid(so->buf))
ReleaseBuffer(so->buf);
MemoryContextDelete(so->tmpCtx); MemoryContextDelete(so->tmpCtx);
pfree(so); pfree(so);

View File

@@ -38,12 +38,12 @@ HnswGetEfConstruction(Relation index)
* Get proc * Get proc
*/ */
FmgrInfo * 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 NULL;
return index_getprocinfo(index, 1, procnum); return index_getprocinfo(rel, 1, procnum);
} }
/* /*
@@ -117,6 +117,7 @@ HnswInitRegisterPage(Relation index, Buffer *buf, Page *page, GenericXLogState *
void void
HnswCommitBuffer(Buffer buf, GenericXLogState *state) HnswCommitBuffer(Buffer buf, GenericXLogState *state)
{ {
MarkBufferDirty(buf);
GenericXLogFinish(state); GenericXLogFinish(state);
UnlockReleaseBuffer(buf); UnlockReleaseBuffer(buf);
} }
@@ -139,21 +140,9 @@ HnswInitNeighbors(HnswElement element, int m)
a = &element->neighbors[lc]; a = &element->neighbors[lc];
a->length = 0; a->length = 0;
a->items = palloc(sizeof(HnswCandidate) * lm); a->items = palloc(sizeof(HnswCandidate) * lm);
a->closerSet = false;
} }
} }
/*
* 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 * Allocate an element
*/ */
@@ -185,8 +174,10 @@ HnswInitElement(ItemPointer heaptid, int m, double ml, int maxLevel)
void void
HnswFreeElement(HnswElement element) HnswFreeElement(HnswElement element)
{ {
HnswFreeNeighbors(element);
list_free_deep(element->heaptids); list_free_deep(element->heaptids);
for (int lc = 0; lc <= element->level; lc++)
pfree(element->neighbors[lc].items);
pfree(element->neighbors);
pfree(element->vec); pfree(element->vec);
pfree(element); pfree(element);
} }
@@ -219,43 +210,25 @@ HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno)
} }
/* /*
* Get the metapage info * Get the entry point
*/ */
void HnswElement
HnswGetMetaPageInfo(Relation index, int *m, HnswElement * entryPoint) HnswGetEntryPoint(Relation index)
{ {
Buffer buf; Buffer buf;
Page page; Page page;
HnswMetaPage metap; HnswMetaPage metap;
HnswElement entryPoint = NULL;
buf = ReadBuffer(index, HNSW_METAPAGE_BLKNO); buf = ReadBuffer(index, HNSW_METAPAGE_BLKNO);
LockBuffer(buf, BUFFER_LOCK_SHARE); LockBuffer(buf, BUFFER_LOCK_SHARE);
page = BufferGetPage(buf); page = BufferGetPage(buf);
metap = HnswPageGetMeta(page); metap = HnswPageGetMeta(page);
if (m != NULL) if (BlockNumberIsValid(metap->entryBlkno))
*m = metap->m; entryPoint = HnswInitElementFromBlock(metap->entryBlkno, metap->entryOffno);
if (entryPoint != NULL)
{
if (BlockNumberIsValid(metap->entryBlkno))
*entryPoint = HnswInitElementFromBlock(metap->entryBlkno, metap->entryOffno);
else
*entryPoint = NULL;
}
UnlockReleaseBuffer(buf); UnlockReleaseBuffer(buf);
}
/*
* Get the entry point
*/
HnswElement
HnswGetEntryPoint(Relation index)
{
HnswElement entryPoint;
HnswGetMetaPageInfo(index, NULL, &entryPoint);
return entryPoint; return entryPoint;
} }
@@ -364,9 +337,10 @@ HnswSetNeighborTuple(HnswNeighborTuple ntup, HnswElement e, int m)
* Load neighbors from page * Load neighbors from page
*/ */
static void 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)); HnswNeighborTuple ntup = (HnswNeighborTuple) PageGetItem(page, PageGetItemId(page, element->neighborOffno));
int m = HnswGetM(index);
int neighborCount = (element->level + 2) * m; int neighborCount = (element->level + 2) * m;
Assert(HnswIsNeighborTuple(ntup)); Assert(HnswIsNeighborTuple(ntup));
@@ -407,7 +381,7 @@ LoadNeighborsFromPage(HnswElement element, Relation index, Page page, int m)
* Load neighbors * Load neighbors
*/ */
void void
HnswLoadNeighbors(HnswElement element, Relation index, int m) HnswLoadNeighbors(HnswElement element, Relation index)
{ {
Buffer buf; Buffer buf;
Page page; Page page;
@@ -416,7 +390,7 @@ HnswLoadNeighbors(HnswElement element, Relation index, int m)
LockBuffer(buf, BUFFER_LOCK_SHARE); LockBuffer(buf, BUFFER_LOCK_SHARE);
page = BufferGetPage(buf); page = BufferGetPage(buf);
LoadNeighborsFromPage(element, index, page, m); LoadNeighborsFromPage(element, index, page);
UnlockReleaseBuffer(buf); UnlockReleaseBuffer(buf);
} }
@@ -569,7 +543,7 @@ AddToVisited(HTAB *v, HnswCandidate * hc, Relation index, bool *found)
* Algorithm 2 from paper * Algorithm 2 from paper
*/ */
List * 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; ListCell *lc2;
@@ -624,7 +598,7 @@ HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *pro
break; break;
if (c->element->neighbors == NULL) if (c->element->neighbors == NULL)
HnswLoadNeighbors(c->element, index, m); HnswLoadNeighbors(c->element, index);
/* Get the neighborhood at layer lc */ /* Get the neighborhood at layer lc */
neighborhood = &c->element->neighbors[lc]; neighborhood = &c->element->neighbors[lc];
@@ -645,7 +619,7 @@ HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *pro
if (index == NULL) if (index == NULL)
eDistance = GetCandidateDistance(e, q, procinfo, collation); eDistance = GetCandidateDistance(e, q, procinfo, collation);
else else
HnswLoadElement(e->element, &eDistance, &q, index, procinfo, collation, inserting); HnswLoadElement(e->element, &eDistance, &q, index, procinfo, collation, loadVec);
Assert(!e->element->deleted); Assert(!e->element->deleted);
@@ -693,34 +667,6 @@ HnswSearchLayer(Datum q, List *ep, int ef, int lc, Relation index, FmgrInfo *pro
return w; return w;
} }
/*
* Compare candidate distances
*/
static int
#if PG_VERSION_NUM >= 130000
CompareCandidateDistances(const ListCell *a, const ListCell *b)
#else
CompareCandidateDistances(const void *a, const void *b)
#endif
{
HnswCandidate *hca = lfirst((ListCell *) a);
HnswCandidate *hcb = lfirst((ListCell *) b);
if (hca->distance < hcb->distance)
return 1;
if (hca->distance > hcb->distance)
return -1;
if (hca->element < hcb->element)
return 1;
if (hca->element > hcb->element)
return -1;
return 0;
}
/* /*
* Calculate the distance between elements * Calculate the distance between elements
*/ */
@@ -777,77 +723,33 @@ CheckElementCloser(HnswCandidate * e, List *r, int lc, FmgrInfo *procinfo, Oid c
* Algorithm 4 from paper * Algorithm 4 from paper
*/ */
static List * static List *
SelectNeighbors(List *c, int m, int lc, FmgrInfo *procinfo, Oid collation, HnswElement e2, HnswCandidate * newCandidate, HnswCandidate * *pruned, bool sortCandidates) SelectNeighbors(List *c, int m, int lc, FmgrInfo *procinfo, Oid collation, HnswCandidate * *pruned)
{ {
List *r = NIL; List *r = NIL;
List *w = list_copy(c); List *w = list_copy(c);
pairingheap *wd; pairingheap *wd;
bool mustCalculate = !e2->neighbors[lc].closerSet;
List *added = NIL;
bool removedAny = false;
if (list_length(w) <= m) if (list_length(w) <= m)
return w; return w;
wd = pairingheap_allocate(CompareNearestCandidates, NULL); wd = pairingheap_allocate(CompareNearestCandidates, NULL);
/* Ensure order of candidates is deterministic for closer caching */
if (sortCandidates)
list_sort(w, CompareCandidateDistances);
while (list_length(w) > 0 && list_length(r) < m) while (list_length(w) > 0 && list_length(r) < m)
{ {
/* Assumes w is already ordered desc */ /* Assumes w is already ordered desc */
HnswCandidate *e = llast(w); HnswCandidate *e = llast(w);
bool closer;
w = list_delete_last(w); w = list_delete_last(w);
/* Use previous state of r and wd to skip work when possible */ closer = CheckElementCloser(e, r, lc, procinfo, collation);
if (mustCalculate)
e->closer = CheckElementCloser(e, r, lc, procinfo, collation);
else if (list_length(added) > 0)
{
/*
* If the current candidate was closer, we only need to compare it
* with the other candidates that we have added.
*/
if (e->closer)
{
e->closer = CheckElementCloser(e, added, lc, procinfo, collation);
if (!e->closer) if (closer)
removedAny = true;
}
else
{
/*
* If we have removed any candidates from closer, a candidate
* that was not closer earlier might now be.
*/
if (removedAny)
{
e->closer = CheckElementCloser(e, r, lc, procinfo, collation);
if (e->closer)
added = lappend(added, e);
}
}
}
else if (e == newCandidate)
{
e->closer = CheckElementCloser(e, r, lc, procinfo, collation);
if (e->closer)
added = lappend(added, e);
}
if (e->closer)
r = lappend(r, e); r = lappend(r, e);
else else
pairingheap_add(wd, &(CreatePairingHeapNode(e)->ph_node)); pairingheap_add(wd, &(CreatePairingHeapNode(e)->ph_node));
} }
/* Cached value can only be used in future if sorted deterministically */
e2->neighbors[lc].closerSet = sortCandidates;
/* Keep pruned connections */ /* Keep pruned connections */
while (!pairingheap_is_empty(wd) && list_length(r) < m) while (!pairingheap_is_empty(wd) && list_length(r) < m)
r = lappend(r, ((HnswPairingHeapNode *) pairingheap_remove_first(wd))->inner); r = lappend(r, ((HnswPairingHeapNode *) pairingheap_remove_first(wd))->inner);
@@ -901,6 +803,28 @@ AddConnections(HnswElement element, List *neighbors, int m, int lc)
a->items[a->length++] = *((HnswCandidate *) lfirst(lc2)); a->items[a->length++] = *((HnswCandidate *) lfirst(lc2));
} }
/*
* Compare candidate distances
*/
static int
#if PG_VERSION_NUM >= 130000
CompareCandidateDistances(const ListCell *a, const ListCell *b)
#else
CompareCandidateDistances(const void *a, const void *b)
#endif
{
HnswCandidate *hca = lfirst((ListCell *) a);
HnswCandidate *hcb = lfirst((ListCell *) b);
if (hca->distance < hcb->distance)
return 1;
if (hca->distance > hcb->distance)
return -1;
return 0;
}
/* /*
* Update connections * Update connections
*/ */
@@ -954,12 +878,13 @@ HnswUpdateConnection(HnswElement element, HnswCandidate * hc, int m, int lc, int
{ {
List *c = NIL; List *c = NIL;
/* Add candidates */ /* Add and sort candidates */
for (int i = 0; i < currentNeighbors->length; i++) for (int i = 0; i < currentNeighbors->length; i++)
c = lappend(c, &currentNeighbors->items[i]); c = lappend(c, &currentNeighbors->items[i]);
c = lappend(c, &hc2); c = lappend(c, &hc2);
list_sort(c, CompareCandidateDistances);
SelectNeighbors(c, m, lc, procinfo, collation, hc->element, &hc2, &pruned, true); SelectNeighbors(c, m, lc, procinfo, collation, &pruned);
/* Should not happen */ /* Should not happen */
if (pruned == NULL) if (pruned == NULL)
@@ -1031,7 +956,7 @@ HnswInsertElement(HnswElement element, HnswElement entryPoint, Relation index, F
/* 1st phase: greedy search to insert level */ /* 1st phase: greedy search to insert level */
for (int lc = entryLevel; lc >= level + 1; lc--) 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; ep = w;
} }
@@ -1049,7 +974,7 @@ HnswInsertElement(HnswElement element, HnswElement entryPoint, Relation index, F
List *neighbors; List *neighbors;
List *lw; 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 */ /* Elements being deleted or skipped can help with search */
/* but should be removed before selecting neighbors */ /* but should be removed before selecting neighbors */
@@ -1058,12 +983,7 @@ HnswInsertElement(HnswElement element, HnswElement entryPoint, Relation index, F
else else
lw = w; lw = w;
/* neighbors = SelectNeighbors(lw, lm, lc, procinfo, collation, NULL);
* Candidates are sorted, but not deterministically. Could set
* sortCandidates to true for in-memory builds to enable closer
* caching, but there does not seem to be a difference in performance.
*/
neighbors = SelectNeighbors(lw, lm, lc, procinfo, collation, element, NULL, NULL, false);
AddConnections(element, neighbors, lm, lc); AddConnections(element, neighbors, lm, lc);

View File

@@ -128,7 +128,10 @@ RemoveHeapTids(HnswVacuumState * vacuumstate)
blkno = HnswPageGetOpaque(page)->nextblkno; blkno = HnswPageGetOpaque(page)->nextblkno;
if (updated) if (updated)
{
MarkBufferDirty(buf);
GenericXLogFinish(state); GenericXLogFinish(state);
}
else else
GenericXLogAbort(state); GenericXLogAbort(state);
@@ -226,6 +229,7 @@ RepairGraphElement(HnswVacuumState * vacuumstate, HnswElement element, HnswEleme
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index)); elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
/* Commit */ /* Commit */
MarkBufferDirty(buf);
GenericXLogFinish(state); GenericXLogFinish(state);
UnlockReleaseBuffer(buf); UnlockReleaseBuffer(buf);
@@ -326,10 +330,7 @@ RepairGraph(HnswVacuumState * vacuumstate)
BufferAccessStrategy bas = vacuumstate->bas; BufferAccessStrategy bas = vacuumstate->bas;
BlockNumber blkno = HNSW_HEAD_BLKNO; BlockNumber blkno = HNSW_HEAD_BLKNO;
/* /* Wait for inserts to complete */
* Wait for inserts to complete. Inserts before this point may have
* neighbors about to be deleted. Inserts after this point will not.
*/
LockPage(index, HNSW_UPDATE_LOCK, ExclusiveLock); LockPage(index, HNSW_UPDATE_LOCK, ExclusiveLock);
UnlockPage(index, HNSW_UPDATE_LOCK, ExclusiveLock); UnlockPage(index, HNSW_UPDATE_LOCK, ExclusiveLock);
@@ -442,11 +443,7 @@ MarkDeleted(HnswVacuumState * vacuumstate)
Relation index = vacuumstate->index; Relation index = vacuumstate->index;
BufferAccessStrategy bas = vacuumstate->bas; BufferAccessStrategy bas = vacuumstate->bas;
/* /* Wait for selects to complete */
* 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.
*/
LockPage(index, HNSW_SCAN_LOCK, ExclusiveLock); LockPage(index, HNSW_SCAN_LOCK, ExclusiveLock);
UnlockPage(index, HNSW_SCAN_LOCK, ExclusiveLock); UnlockPage(index, HNSW_SCAN_LOCK, ExclusiveLock);
@@ -543,6 +540,9 @@ MarkDeleted(HnswVacuumState * vacuumstate)
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index)); elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
/* Commit */ /* Commit */
MarkBufferDirty(buf);
if (nbuf != buf)
MarkBufferDirty(nbuf);
GenericXLogFinish(state); GenericXLogFinish(state);
if (nbuf != buf) if (nbuf != buf)
UnlockReleaseBuffer(nbuf); UnlockReleaseBuffer(nbuf);
@@ -582,6 +582,7 @@ InitVacuumState(HnswVacuumState * vacuumstate, IndexVacuumInfo *info, IndexBulkD
vacuumstate->stats = stats; vacuumstate->stats = stats;
vacuumstate->callback = callback; vacuumstate->callback = callback;
vacuumstate->callback_state = callback_state; vacuumstate->callback_state = callback_state;
vacuumstate->m = HnswGetM(index);
vacuumstate->efConstruction = HnswGetEfConstruction(index); vacuumstate->efConstruction = HnswGetEfConstruction(index);
vacuumstate->bas = GetAccessStrategy(BAS_BULKREAD); vacuumstate->bas = GetAccessStrategy(BAS_BULKREAD);
vacuumstate->procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC); vacuumstate->procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC);
@@ -591,9 +592,6 @@ InitVacuumState(HnswVacuumState * vacuumstate, IndexVacuumInfo *info, IndexBulkD
"Hnsw vacuum temporary context", "Hnsw vacuum temporary context",
ALLOCSET_DEFAULT_SIZES); ALLOCSET_DEFAULT_SIZES);
/* Get m from metapage */
HnswGetMetaPageInfo(index, &vacuumstate->m, NULL);
/* Create hash table */ /* Create hash table */
hash_ctl.keysize = sizeof(ItemPointerData); hash_ctl.keysize = sizeof(ItemPointerData);
hash_ctl.entrysize = sizeof(ItemPointerData); hash_ctl.entrysize = sizeof(ItemPointerData);

View File

@@ -287,7 +287,7 @@ InsertTuples(Relation index, IvfflatBuildState * buildstate, ForkNumber forkNum)
#else #else
TupleTableSlot *slot = MakeSingleTupleTableSlot(buildstate->tupdesc); TupleTableSlot *slot = MakeSingleTupleTableSlot(buildstate->tupdesc);
#endif #endif
TupleDesc tupdesc = IvfflatTupleDesc(index); TupleDesc tupdesc = RelationGetDescr(index);
UpdateProgress(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_IVFFLAT_PHASE_LOAD); UpdateProgress(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_IVFFLAT_PHASE_LOAD);
@@ -339,8 +339,6 @@ InsertTuples(Relation index, IvfflatBuildState * buildstate, ForkNumber forkNum)
/* Set the start and insert pages */ /* Set the start and insert pages */
IvfflatUpdateList(index, buildstate->listInfo[i], insertPage, InvalidBlockNumber, startPage, forkNum); IvfflatUpdateList(index, buildstate->listInfo[i], insertPage, InvalidBlockNumber, startPage, forkNum);
} }
FreeTupleDesc(tupdesc);
} }
/* /*
@@ -508,30 +506,29 @@ CreateListPages(Relation index, VectorArray centers, int dimensions,
Buffer buf; Buffer buf;
Page page; Page page;
GenericXLogState *state; GenericXLogState *state;
Size listSize; OffsetNumber offno;
Size itemsz;
IvfflatList list; IvfflatList list;
listSize = MAXALIGN(IVFFLAT_LIST_SIZE(dimensions)); itemsz = MAXALIGN(IVFFLAT_LIST_SIZE(dimensions));
list = palloc(listSize); list = palloc(itemsz);
buf = IvfflatNewBuffer(index, forkNum); buf = IvfflatNewBuffer(index, forkNum);
IvfflatInitRegisterPage(index, &buf, &page, &state); IvfflatInitRegisterPage(index, &buf, &page, &state);
for (int i = 0; i < lists; i++) for (int i = 0; i < lists; i++)
{ {
OffsetNumber offno;
/* Load list */ /* Load list */
list->startPage = InvalidBlockNumber; list->startPage = InvalidBlockNumber;
list->insertPage = InvalidBlockNumber; list->insertPage = InvalidBlockNumber;
memcpy(&list->center, VectorArrayGet(centers, i), VECTOR_SIZE(dimensions)); memcpy(&list->center, VectorArrayGet(centers, i), VECTOR_SIZE(dimensions));
/* Ensure free space */ /* Ensure free space */
if (PageGetFreeSpace(page) < listSize) if (PageGetFreeSpace(page) < itemsz)
IvfflatAppendPage(index, &buf, &page, &state, forkNum); IvfflatAppendPage(index, &buf, &page, &state, forkNum);
/* Add the item */ /* Add the item */
offno = PageAddItem(page, (Item) list, listSize, InvalidOffsetNumber, false, false); offno = PageAddItem(page, (Item) list, itemsz, InvalidOffsetNumber, false, false);
if (offno == InvalidOffsetNumber) if (offno == InvalidOffsetNumber)
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index)); 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; int lists;
double ratio; double ratio;
double spc_seq_page_cost; double spc_seq_page_cost;
Relation index; Relation indexRel;
#if PG_VERSION_NUM < 120000 #if PG_VERSION_NUM < 120000
List *qinfos; List *qinfos;
#endif #endif
@@ -89,9 +89,9 @@ ivfflatcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
MemSet(&costs, 0, sizeof(costs)); MemSet(&costs, 0, sizeof(costs));
index = index_open(path->indexinfo->indexoid, NoLock); indexRel = index_open(path->indexinfo->indexoid, NoLock);
IvfflatGetMetaPageInfo(index, &lists, NULL); lists = IvfflatGetLists(indexRel);
index_close(index, NoLock); index_close(indexRel, NoLock);
/* Get the ratio of lists that we need to visit */ /* Get the ratio of lists that we need to visit */
ratio = ((double) ivfflat_probes) / lists; ratio = ((double) ivfflat_probes) / lists;

View File

@@ -244,8 +244,8 @@ typedef struct IvfflatScanList
typedef struct IvfflatScanOpaqueData typedef struct IvfflatScanOpaqueData
{ {
int probes; int probes;
int dimensions;
bool first; bool first;
Buffer buf;
/* Sorting */ /* Sorting */
Tuplesortstate *sortstate; Tuplesortstate *sortstate;
@@ -275,10 +275,9 @@ VectorArray VectorArrayInit(int maxlen, int dimensions);
void VectorArrayFree(VectorArray arr); void VectorArrayFree(VectorArray arr);
void PrintVectorArray(char *msg, VectorArray arr); void PrintVectorArray(char *msg, VectorArray arr);
void IvfflatKmeans(Relation index, VectorArray samples, VectorArray centers); 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); bool IvfflatNormValue(FmgrInfo *procinfo, Oid collation, Datum *value, Vector * result);
int IvfflatGetLists(Relation index); 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 IvfflatUpdateList(Relation index, ListInfo listInfo, BlockNumber insertPage, BlockNumber originalInsertPage, BlockNumber startPage, ForkNumber forkNum);
void IvfflatCommitBuffer(Buffer buf, GenericXLogState *state); void IvfflatCommitBuffer(Buffer buf, GenericXLogState *state);
void IvfflatAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **state, ForkNumber forkNum); void IvfflatAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **state, ForkNumber forkNum);
@@ -287,7 +286,6 @@ void IvfflatInitPage(Buffer buf, Page page);
void IvfflatInitRegisterPage(Relation index, Buffer *buf, Page *page, GenericXLogState **state); void IvfflatInitRegisterPage(Relation index, Buffer *buf, Page *page, GenericXLogState **state);
void IvfflatInit(void); void IvfflatInit(void);
PGDLLEXPORT void IvfflatParallelBuildMain(dsm_segment *seg, shm_toc *toc); PGDLLEXPORT void IvfflatParallelBuildMain(dsm_segment *seg, shm_toc *toc);
TupleDesc IvfflatTupleDesc(Relation index);
/* Index access methods */ /* Index access methods */
IndexBuildResult *ivfflatbuild(Relation heap, Relation index, IndexInfo *indexInfo); IndexBuildResult *ivfflatbuild(Relation heap, Relation index, IndexInfo *indexInfo);

View File

@@ -11,37 +11,36 @@
* Find the list that minimizes the distance function * Find the list that minimizes the distance function
*/ */
static void 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; double minDistance = DBL_MAX;
BlockNumber nextblkno = IVFFLAT_HEAD_BLKNO; BlockNumber nextblkno = IVFFLAT_HEAD_BLKNO;
FmgrInfo *procinfo; FmgrInfo *procinfo;
Oid collation; Oid collation;
OffsetNumber offno;
OffsetNumber maxoffno;
/* Avoid compiler warning */ /* Avoid compiler warning */
listInfo->blkno = nextblkno; listInfo->blkno = nextblkno;
listInfo->offno = FirstOffsetNumber; listInfo->offno = FirstOffsetNumber;
procinfo = index_getprocinfo(index, 1, IVFFLAT_DISTANCE_PROC); procinfo = index_getprocinfo(rel, 1, IVFFLAT_DISTANCE_PROC);
collation = index->rd_indcollation[0]; collation = rel->rd_indcollation[0];
/* Search all list pages */ /* Search all list pages */
while (BlockNumberIsValid(nextblkno)) while (BlockNumberIsValid(nextblkno))
{ {
Buffer cbuf; cbuf = ReadBuffer(rel, nextblkno);
Page cpage;
OffsetNumber maxoffno;
cbuf = ReadBuffer(index, nextblkno);
LockBuffer(cbuf, BUFFER_LOCK_SHARE); LockBuffer(cbuf, BUFFER_LOCK_SHARE);
cpage = BufferGetPage(cbuf); cpage = BufferGetPage(cbuf);
maxoffno = PageGetMaxOffsetNumber(cpage); 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)); list = (IvfflatList) PageGetItem(cpage, PageGetItemId(cpage, offno));
distance = DatumGetFloat8(FunctionCall2Coll(procinfo, collation, values[0], PointerGetDatum(&list->center))); 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 * Insert a tuple into the index
*/ */
static void 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; IndexTuple itup;
Datum value; Datum value;
@@ -81,33 +80,33 @@ InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, R
value = PointerGetDatum(PG_DETOAST_DATUM(values[0])); value = PointerGetDatum(PG_DETOAST_DATUM(values[0]));
/* Normalize if needed */ /* Normalize if needed */
normprocinfo = IvfflatOptionalProcInfo(index, IVFFLAT_NORM_PROC); normprocinfo = IvfflatOptionalProcInfo(rel, IVFFLAT_NORM_PROC);
if (normprocinfo != NULL) if (normprocinfo != NULL)
{ {
if (!IvfflatNormValue(normprocinfo, index->rd_indcollation[0], &value, NULL)) if (!IvfflatNormValue(normprocinfo, rel->rd_indcollation[0], &value, NULL))
return; return;
} }
/* Find the insert page - sets the page and list info */ /* Find the insert page - sets the page and list info */
FindInsertPage(index, values, &insertPage, &listInfo); FindInsertPage(rel, values, &insertPage, &listInfo);
Assert(BlockNumberIsValid(insertPage)); Assert(BlockNumberIsValid(insertPage));
originalInsertPage = insertPage; originalInsertPage = insertPage;
/* Form tuple */ /* Form tuple */
itup = index_form_tuple(IvfflatTupleDesc(index), &value, isnull); itup = index_form_tuple(RelationGetDescr(rel), &value, isnull);
itup->t_tid = *heap_tid; itup->t_tid = *heap_tid;
/* Get tuple size */ /* Get tuple size */
itemsz = MAXALIGN(IndexTupleSize(itup)); itemsz = MAXALIGN(IndexTupleSize(itup));
Assert(itemsz <= BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(IvfflatPageOpaqueData)) - sizeof(ItemIdData)); Assert(itemsz <= BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(IvfflatPageOpaqueData)));
/* Find a page to insert the item */ /* Find a page to insert the item */
for (;;) for (;;)
{ {
buf = ReadBuffer(index, insertPage); buf = ReadBuffer(rel, insertPage);
LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE); LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
state = GenericXLogStart(index); state = GenericXLogStart(rel);
page = GenericXLogRegisterBuffer(state, buf, 0); page = GenericXLogRegisterBuffer(state, buf, 0);
if (PageGetFreeSpace(page) >= itemsz) if (PageGetFreeSpace(page) >= itemsz)
@@ -127,9 +126,9 @@ InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, R
Page newpage; Page newpage;
/* Add a new page */ /* Add a new page */
LockRelationForExtension(index, ExclusiveLock); LockRelationForExtension(rel, ExclusiveLock);
newbuf = IvfflatNewBuffer(index, MAIN_FORKNUM); newbuf = IvfflatNewBuffer(rel, MAIN_FORKNUM);
UnlockRelationForExtension(index, ExclusiveLock); UnlockRelationForExtension(rel, ExclusiveLock);
/* Init new page */ /* Init new page */
newpage = GenericXLogRegisterBuffer(state, newbuf, GENERIC_XLOG_FULL_IMAGE); newpage = GenericXLogRegisterBuffer(state, newbuf, GENERIC_XLOG_FULL_IMAGE);
@@ -142,13 +141,15 @@ InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, R
IvfflatPageGetOpaque(page)->nextblkno = insertPage; IvfflatPageGetOpaque(page)->nextblkno = insertPage;
/* Commit */ /* Commit */
MarkBufferDirty(newbuf);
MarkBufferDirty(buf);
GenericXLogFinish(state); GenericXLogFinish(state);
/* Unlock previous buffer */ /* Unlock previous buffer */
UnlockReleaseBuffer(buf); UnlockReleaseBuffer(buf);
/* Prepare new buffer */ /* Prepare new buffer */
state = GenericXLogStart(index); state = GenericXLogStart(rel);
buf = newbuf; buf = newbuf;
page = GenericXLogRegisterBuffer(state, buf, 0); page = GenericXLogRegisterBuffer(state, buf, 0);
break; break;
@@ -157,13 +158,13 @@ InsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, R
/* Add to next offset */ /* Add to next offset */
if (PageAddItem(page, (Item) itup, itemsz, InvalidOffsetNumber, false, false) == InvalidOffsetNumber) 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); IvfflatCommitBuffer(buf, state);
/* Update the insert page */ /* Update the insert page */
if (insertPage != originalInsertPage) 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; FmgrInfo *procinfo;
Oid collation; Oid collation;
int64 j; int64 j;
double distance;
double sum;
double choice;
Vector *vec;
float *weight = palloc(samples->length * sizeof(float)); float *weight = palloc(samples->length * sizeof(float));
int numCenters = centers->maxlen; int numCenters = centers->maxlen;
int numSamples = samples->length; int numSamples = samples->length;
@@ -29,21 +33,17 @@ InitCenters(Relation index, VectorArray samples, VectorArray centers, float *low
centers->length++; centers->length++;
for (j = 0; j < numSamples; j++) for (j = 0; j < numSamples; j++)
weight[j] = FLT_MAX; weight[j] = DBL_MAX;
for (int i = 0; i < numCenters; i++) for (int i = 0; i < numCenters; i++)
{ {
double sum;
double choice;
CHECK_FOR_INTERRUPTS(); CHECK_FOR_INTERRUPTS();
sum = 0.0; sum = 0.0;
for (j = 0; j < numSamples; j++) for (j = 0; j < numSamples; j++)
{ {
Vector *vec = VectorArrayGet(samples, j); vec = VectorArrayGet(samples, j);
double distance;
/* Only need to compute distance for new center */ /* Only need to compute distance for new center */
/* TODO Use triangle inequality to reduce distance calculations */ /* TODO Use triangle inequality to reduce distance calculations */
@@ -112,6 +112,7 @@ CompareVectors(const void *a, const void *b)
static void static void
QuickCenters(Relation index, VectorArray samples, VectorArray centers) QuickCenters(Relation index, VectorArray samples, VectorArray centers)
{ {
Vector *vec;
int dimensions = centers->dim; int dimensions = centers->dim;
Oid collation = index->rd_indcollation[0]; Oid collation = index->rd_indcollation[0];
FmgrInfo *normprocinfo = IvfflatOptionalProcInfo(index, IVFFLAT_KMEANS_NORM_PROC); 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); qsort(samples->items, samples->length, VECTOR_SIZE(samples->dim), CompareVectors);
for (int i = 0; i < samples->length; i++) 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) 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 */ /* Fill remaining with random data */
while (centers->length < centers->maxlen) while (centers->length < centers->maxlen)
{ {
Vector *vec = VectorArrayGet(centers, centers->length); vec = VectorArrayGet(centers, centers->length);
SET_VARSIZE(vec, VECTOR_SIZE(dimensions)); SET_VARSIZE(vec, VECTOR_SIZE(dimensions));
vec->dim = dimensions; vec->dim = dimensions;
@@ -167,6 +168,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
Oid collation; Oid collation;
Vector *vec; Vector *vec;
Vector *newCenter; Vector *newCenter;
int iteration;
int64 j; int64 j;
int64 k; int64 k;
int dimensions = centers->dim; int dimensions = centers->dim;
@@ -180,6 +182,14 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
float *s; float *s;
float *halfcdist; float *halfcdist;
float *newcdist; float *newcdist;
int changes;
double minDistance;
int closestCenter;
double distance;
bool rj;
bool rjreset;
double dxcx;
double dxc;
/* Calculate allocation sizes */ /* Calculate allocation sizes */
Size samplesSize = VECTOR_ARRAY_SIZE(samples->maxlen, samples->dim); 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) */ /* Assign each x to its closest initial center c(x) = argmin d(x,c) */
for (j = 0; j < numSamples; j++) for (j = 0; j < numSamples; j++)
{ {
float minDistance = FLT_MAX; minDistance = DBL_MAX;
int closestCenter = 0; closestCenter = 0;
/* Find closest center */ /* Find closest center */
for (k = 0; k < numCenters; k++) for (k = 0; k < numCenters; k++)
{ {
/* TODO Use Lemma 1 in k-means++ initialization */ /* TODO Use Lemma 1 in k-means++ initialization */
float distance = lowerBound[j * numCenters + k]; distance = lowerBound[j * numCenters + k];
if (distance < minDistance) if (distance < minDistance)
{ {
@@ -258,14 +268,13 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
} }
/* Give 500 iterations to converge */ /* 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 */ /* Can take a while, so ensure we can interrupt */
CHECK_FOR_INTERRUPTS(); CHECK_FOR_INTERRUPTS();
changes = 0;
/* Step 1: For all centers, compute distance */ /* Step 1: For all centers, compute distance */
for (j = 0; j < numCenters; j++) for (j = 0; j < numCenters; j++)
{ {
@@ -273,8 +282,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
for (k = j + 1; k < numCenters; k++) 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[j * numCenters + k] = distance;
halfcdist[k * numCenters + j] = 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 all centers c, compute s(c) */
for (j = 0; j < numCenters; j++) for (j = 0; j < numCenters; j++)
{ {
float minDistance = FLT_MAX; minDistance = DBL_MAX;
for (k = 0; k < numCenters; k++) for (k = 0; k < numCenters; k++)
{ {
float distance;
if (j == k) if (j == k)
continue; continue;
@@ -304,8 +310,6 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
for (j = 0; j < numSamples; j++) for (j = 0; j < numSamples; j++)
{ {
bool rj;
/* Step 2: Identify all points x such that u(x) <= s(c(x)) */ /* Step 2: Identify all points x such that u(x) <= s(c(x)) */
if (upperBound[j] <= s[closestCenters[j]]) if (upperBound[j] <= s[closestCenters[j]])
continue; continue;
@@ -314,8 +318,6 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
for (k = 0; k < numCenters; k++) for (k = 0; k < numCenters; k++)
{ {
float dxcx;
/* Step 3: For all remaining points x and centers c */ /* Step 3: For all remaining points x and centers c */
if (k == closestCenters[j]) if (k == closestCenters[j])
continue; continue;
@@ -345,7 +347,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
/* Step 3b */ /* Step 3b */
if (dxcx > lowerBound[j * numCenters + k] || dxcx > halfcdist[closestCenters[j] * numCenters + k]) 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 */ /* d(x,c) calculated */
lowerBound[j * numCenters + k] = dxc; lowerBound[j * numCenters + k] = dxc;
@@ -359,6 +361,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
changes++; changes++;
} }
} }
} }
} }
@@ -375,8 +378,6 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
for (j = 0; j < numSamples; j++) for (j = 0; j < numSamples; j++)
{ {
int closestCenter;
vec = VectorArrayGet(samples, j); vec = VectorArrayGet(samples, j);
closestCenter = closestCenters[j]; closestCenter = closestCenters[j];
@@ -425,7 +426,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
{ {
for (k = 0; k < numCenters; k++) for (k = 0; k < numCenters; k++)
{ {
float distance = lowerBound[j * numCenters + k] - newcdist[k]; distance = lowerBound[j * numCenters + k] - newcdist[k];
if (distance < 0) if (distance < 0)
distance = 0; distance = 0;
@@ -441,7 +442,7 @@ ElkanKmeans(Relation index, VectorArray samples, VectorArray centers)
/* Step 7 */ /* Step 7 */
for (j = 0; j < numCenters; j++) 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) if (changes == 0 && iteration != 0)
break; break;
@@ -464,6 +465,9 @@ static void
CheckCenters(Relation index, VectorArray centers) CheckCenters(Relation index, VectorArray centers)
{ {
FmgrInfo *normprocinfo; FmgrInfo *normprocinfo;
Oid collation;
Vector *vec;
double norm;
if (centers->length != centers->maxlen) if (centers->length != centers->maxlen)
elog(ERROR, "Not enough centers. Please report a bug."); elog(ERROR, "Not enough centers. Please report a bug.");
@@ -471,7 +475,7 @@ CheckCenters(Relation index, VectorArray centers)
/* Ensure no NaN or infinite values */ /* Ensure no NaN or infinite values */
for (int i = 0; i < centers->length; i++) 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++) for (int j = 0; j < vec->dim; j++)
{ {
@@ -497,12 +501,11 @@ CheckCenters(Relation index, VectorArray centers)
normprocinfo = IvfflatOptionalProcInfo(index, IVFFLAT_NORM_PROC); normprocinfo = IvfflatOptionalProcInfo(index, IVFFLAT_NORM_PROC);
if (normprocinfo != NULL) if (normprocinfo != NULL)
{ {
Oid collation = index->rd_indcollation[0]; collation = index->rd_indcollation[0];
for (int i = 0; i < centers->length; i++) 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) if (norm == 0)
elog(ERROR, "Zero norm detected. Please report a bug."); 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 static void
GetScanLists(IndexScanDesc scan, Datum value) 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; BlockNumber nextblkno = IVFFLAT_HEAD_BLKNO;
int listCount = 0; int listCount = 0;
IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
double distance;
IvfflatScanList *scanlist;
double maxDistance = DBL_MAX; double maxDistance = DBL_MAX;
/* Search all list pages */ /* Search all list pages */
while (BlockNumberIsValid(nextblkno)) while (BlockNumberIsValid(nextblkno))
{ {
Buffer cbuf;
Page cpage;
OffsetNumber maxoffno;
cbuf = ReadBuffer(scan->indexRelation, nextblkno); cbuf = ReadBuffer(scan->indexRelation, nextblkno);
LockBuffer(cbuf, BUFFER_LOCK_SHARE); LockBuffer(cbuf, BUFFER_LOCK_SHARE);
cpage = BufferGetPage(cbuf); cpage = BufferGetPage(cbuf);
maxoffno = PageGetMaxOffsetNumber(cpage); 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)); list = (IvfflatList) PageGetItem(cpage, PageGetItemId(cpage, offno));
double distance;
/* Use procinfo from the index instead of scan key for performance */ /* Use procinfo from the index instead of scan key for performance */
distance = DatumGetFloat8(FunctionCall2Coll(so->procinfo, so->collation, PointerGetDatum(&list->center), value)); distance = DatumGetFloat8(FunctionCall2Coll(so->procinfo, so->collation, PointerGetDatum(&list->center), value));
if (listCount < so->probes) if (listCount < so->probes)
{ {
IvfflatScanList *scanlist;
scanlist = &so->lists[listCount]; scanlist = &so->lists[listCount];
scanlist->startPage = list->startPage; scanlist->startPage = list->startPage;
scanlist->distance = distance; scanlist->distance = distance;
@@ -75,8 +75,6 @@ GetScanLists(IndexScanDesc scan, Datum value)
} }
else if (distance < maxDistance) else if (distance < maxDistance)
{ {
IvfflatScanList *scanlist;
/* Remove */ /* Remove */
scanlist = (IvfflatScanList *) pairingheap_remove_first(so->listQueue); scanlist = (IvfflatScanList *) pairingheap_remove_first(so->listQueue);
@@ -103,6 +101,14 @@ static void
GetScanItems(IndexScanDesc scan, Datum value) GetScanItems(IndexScanDesc scan, Datum value)
{ {
IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque; 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); TupleDesc tupdesc = RelationGetDescr(scan->indexRelation);
double tuples = 0; double tuples = 0;
@@ -122,28 +128,19 @@ GetScanItems(IndexScanDesc scan, Datum value)
/* Search closest probes lists */ /* Search closest probes lists */
while (!pairingheap_is_empty(so->listQueue)) 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 */ /* Search all entry pages for list */
while (BlockNumberIsValid(searchPage)) while (BlockNumberIsValid(searchPage))
{ {
Buffer buf;
Page page;
OffsetNumber maxoffno;
buf = ReadBufferExtended(scan->indexRelation, MAIN_FORKNUM, searchPage, RBM_NORMAL, bas); buf = ReadBufferExtended(scan->indexRelation, MAIN_FORKNUM, searchPage, RBM_NORMAL, bas);
LockBuffer(buf, BUFFER_LOCK_SHARE); LockBuffer(buf, BUFFER_LOCK_SHARE);
page = BufferGetPage(buf); page = BufferGetPage(buf);
maxoffno = PageGetMaxOffsetNumber(page); maxoffno = PageGetMaxOffsetNumber(page);
for (OffsetNumber offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno)) for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
{ {
IndexTuple itup; itup = (IndexTuple) PageGetItem(page, PageGetItemId(page, offno));
Datum datum;
bool isnull;
ItemId itemid = PageGetItemId(page, offno);
itup = (IndexTuple) PageGetItem(page, itemid);
datum = index_getattr(itup, 1, tupdesc, &isnull); datum = index_getattr(itup, 1, tupdesc, &isnull);
/* /*
@@ -157,6 +154,8 @@ GetScanItems(IndexScanDesc scan, Datum value)
slot->tts_isnull[0] = false; slot->tts_isnull[0] = false;
slot->tts_values[1] = PointerGetDatum(&itup->t_tid); slot->tts_values[1] = PointerGetDatum(&itup->t_tid);
slot->tts_isnull[1] = false; slot->tts_isnull[1] = false;
slot->tts_values[2] = Int32GetDatum((int) searchPage);
slot->tts_isnull[2] = false;
ExecStoreVirtualTuple(slot); ExecStoreVirtualTuple(slot);
tuplesort_puttupleslot(so->sortstate, slot); tuplesort_puttupleslot(so->sortstate, slot);
@@ -181,6 +180,29 @@ GetScanItems(IndexScanDesc scan, Datum value)
tuplesort_performsort(so->sortstate); tuplesort_performsort(so->sortstate);
} }
/*
* Get dimensions from metapage
*/
static int
GetDimensions(Relation index)
{
Buffer buf;
Page page;
IvfflatMetaPage metap;
int dimensions;
buf = ReadBuffer(index, IVFFLAT_METAPAGE_BLKNO);
LockBuffer(buf, BUFFER_LOCK_SHARE);
page = BufferGetPage(buf);
metap = IvfflatPageGetMeta(page);
dimensions = metap->dimensions;
UnlockReleaseBuffer(buf);
return dimensions;
}
/* /*
* Prepare for an index scan * Prepare for an index scan
*/ */
@@ -190,7 +212,6 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
IndexScanDesc scan; IndexScanDesc scan;
IvfflatScanOpaque so; IvfflatScanOpaque so;
int lists; int lists;
int dimensions;
AttrNumber attNums[] = {1}; AttrNumber attNums[] = {1};
Oid sortOperators[] = {Float8LessOperator}; Oid sortOperators[] = {Float8LessOperator};
Oid sortCollations[] = {InvalidOid}; Oid sortCollations[] = {InvalidOid};
@@ -198,17 +219,15 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
int probes = ivfflat_probes; int probes = ivfflat_probes;
scan = RelationGetIndexScan(index, nkeys, norderbys); scan = RelationGetIndexScan(index, nkeys, norderbys);
lists = IvfflatGetLists(scan->indexRelation);
/* Get lists and dimensions from metapage */
IvfflatGetMetaPageInfo(index, &lists, &dimensions);
if (probes > lists) if (probes > lists)
probes = lists; probes = lists;
so = (IvfflatScanOpaque) palloc(offsetof(IvfflatScanOpaqueData, lists) + probes * sizeof(IvfflatScanList)); so = (IvfflatScanOpaque) palloc(offsetof(IvfflatScanOpaqueData, lists) + probes * sizeof(IvfflatScanList));
so->buf = InvalidBuffer;
so->first = true; so->first = true;
so->probes = probes; so->probes = probes;
so->dimensions = dimensions;
/* Set support functions */ /* Set support functions */
so->procinfo = index_getprocinfo(index, 1, IVFFLAT_DISTANCE_PROC); so->procinfo = index_getprocinfo(index, 1, IVFFLAT_DISTANCE_PROC);
@@ -217,12 +236,13 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
/* Create tuple description for sorting */ /* Create tuple description for sorting */
#if PG_VERSION_NUM >= 120000 #if PG_VERSION_NUM >= 120000
so->tupdesc = CreateTemplateTupleDesc(2); so->tupdesc = CreateTemplateTupleDesc(3);
#else #else
so->tupdesc = CreateTemplateTupleDesc(2, false); so->tupdesc = CreateTemplateTupleDesc(3, false);
#endif #endif
TupleDescInitEntry(so->tupdesc, (AttrNumber) 1, "distance", FLOAT8OID, -1, 0); 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 */ /* Prep sort */
so->sortstate = tuplesort_begin_heap(so->tupdesc, 1, attNums, sortOperators, sortCollations, nullsFirstFlags, work_mem, NULL, false); so->sortstate = tuplesort_begin_heap(so->tupdesc, 1, attNums, sortOperators, sortCollations, nullsFirstFlags, work_mem, NULL, false);
@@ -288,13 +308,8 @@ ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
if (scan->orderByData == NULL) if (scan->orderByData == NULL)
elog(ERROR, "cannot scan ivfflat index without order"); elog(ERROR, "cannot scan ivfflat index without order");
/* Requires MVCC-compliant snapshot as not able to pin during sorting */
/* https://www.postgresql.org/docs/current/index-locking.html */
if (!IsMVCCSnapshot(scan->xs_snapshot))
elog(ERROR, "non-MVCC snapshots are not supported with ivfflat");
if (scan->orderByData->sk_flags & SK_ISNULL) if (scan->orderByData->sk_flags & SK_ISNULL)
value = PointerGetDatum(InitVector(so->dimensions)); value = PointerGetDatum(InitVector(GetDimensions(scan->indexRelation)));
else else
{ {
value = scan->orderByData->sk_argument; value = scan->orderByData->sk_argument;
@@ -319,14 +334,27 @@ ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
if (tuplesort_gettupleslot(so->sortstate, true, false, so->slot, NULL)) 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 #if PG_VERSION_NUM >= 120000
scan->xs_heaptid = *heaptid; scan->xs_heaptid = *tid;
#else #else
scan->xs_ctup.t_self = *heaptid; scan->xs_ctup.t_self = *tid;
#endif #endif
/* Unpin buffer */
if (BufferIsValid(so->buf))
ReleaseBuffer(so->buf);
/*
* 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; scan->xs_recheckorderby = false;
return true; return true;
} }
@@ -342,6 +370,10 @@ ivfflatendscan(IndexScanDesc scan)
{ {
IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque; IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
/* Release pin */
if (BufferIsValid(so->buf))
ReleaseBuffer(so->buf);
pairingheap_free(so->listQueue); pairingheap_free(so->listQueue);
tuplesort_end(so->sortstate); tuplesort_end(so->sortstate);

View File

@@ -4,10 +4,6 @@
#include "storage/bufmgr.h" #include "storage/bufmgr.h"
#include "vector.h" #include "vector.h"
#if PG_VERSION_NUM < 130000
#define TYPSTORAGE_PLAIN 'p'
#endif
/* /*
* Allocate a vector array * Allocate a vector array
*/ */
@@ -61,12 +57,12 @@ IvfflatGetLists(Relation index)
* Get proc * Get proc
*/ */
FmgrInfo * 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 NULL;
return index_getprocinfo(index, 1, procnum); return index_getprocinfo(rel, 1, procnum);
} }
/* /*
@@ -140,6 +136,7 @@ IvfflatInitRegisterPage(Relation index, Buffer *buf, Page *page, GenericXLogStat
void void
IvfflatCommitBuffer(Buffer buf, GenericXLogState *state) IvfflatCommitBuffer(Buffer buf, GenericXLogState *state)
{ {
MarkBufferDirty(buf);
GenericXLogFinish(state); GenericXLogFinish(state);
UnlockReleaseBuffer(buf); UnlockReleaseBuffer(buf);
} }
@@ -163,6 +160,8 @@ IvfflatAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **st
IvfflatInitPage(newbuf, newpage); IvfflatInitPage(newbuf, newpage);
/* Commit */ /* Commit */
MarkBufferDirty(*buf);
MarkBufferDirty(newbuf);
GenericXLogFinish(*state); GenericXLogFinish(*state);
/* Unlock */ /* Unlock */
@@ -173,29 +172,6 @@ IvfflatAppendPage(Relation index, Buffer *buf, Page *page, GenericXLogState **st
*buf = newbuf; *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 * Update the start or insert page of a list
*/ */
@@ -242,17 +218,3 @@ IvfflatUpdateList(Relation index, ListInfo listInfo,
UnlockReleaseBuffer(buf); UnlockReleaseBuffer(buf);
} }
} }
/*
* Get the tuple descriptor
*/
TupleDesc
IvfflatTupleDesc(Relation index)
{
TupleDesc tupdesc = CreateTupleDescCopyConstr(RelationGetDescr(index));
/* Prevent compression */
TupleDescAttr(tupdesc, 0)->attstorage = TYPSTORAGE_PLAIN;
return tupdesc;
}

View File

@@ -107,6 +107,7 @@ ivfflatbulkdelete(IndexVacuumInfo *info, IndexBulkDeleteResult *stats,
{ {
/* Delete tuples */ /* Delete tuples */
PageIndexMultiDelete(page, deletable, ndeletable); PageIndexMultiDelete(page, deletable, ndeletable);
MarkBufferDirty(buf);
GenericXLogFinish(state); GenericXLogFinish(state);
} }
else else

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 = '0.5.1' default_version = '0.5.0'
module_pathname = '$libdir/vector' module_pathname = '$libdir/vector'
relocatable = true relocatable = true