mirror of
https://github.com/pgvector/pgvector.git
synced 2026-07-22 03:57:34 +08:00
Compare commits
1 Commits
hnsw-array
...
non-mvcc-s
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
cb91de3332 |
2
Makefile
2
Makefile
@@ -3,7 +3,7 @@ EXTVERSION = 0.5.0
|
||||
|
||||
MODULE_big = vector
|
||||
DATA = $(wildcard sql/*--*.sql)
|
||||
OBJS = src/float4.o src/hnsw.o src/hnswbuild.o src/hnswinsert.o src/hnswscan.o src/hnswutils.o src/hnswvacuum.o src/ivfbuild.o src/ivfflat.o src/ivfinsert.o src/ivfkmeans.o src/ivfscan.o src/ivfutils.o src/ivfvacuum.o src/vector.o
|
||||
OBJS = src/hnsw.o src/hnswbuild.o src/hnswinsert.o src/hnswscan.o src/hnswutils.o src/hnswvacuum.o src/ivfbuild.o src/ivfflat.o src/ivfinsert.o src/ivfkmeans.o src/ivfscan.o src/ivfutils.o src/ivfvacuum.o src/vector.o
|
||||
HEADERS = src/vector.h
|
||||
|
||||
TESTS = $(wildcard test/sql/*.sql)
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
EXTENSION = vector
|
||||
EXTVERSION = 0.5.0
|
||||
|
||||
OBJS = src\float4.obj 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
|
||||
|
||||
REGRESS = btree cast copy functions input ivfflat_cosine ivfflat_ip ivfflat_l2 ivfflat_options ivfflat_unlogged
|
||||
|
||||
@@ -34,9 +34,6 @@ CREATE TYPE vector (
|
||||
CREATE FUNCTION l2_distance(vector, vector) RETURNS float8
|
||||
AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
|
||||
|
||||
CREATE FUNCTION l2_distance(float4[], float4[]) RETURNS float8
|
||||
AS 'MODULE_PATHNAME', 'float4_l2_distance' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
|
||||
|
||||
CREATE FUNCTION inner_product(vector, vector) RETURNS float8
|
||||
AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
|
||||
|
||||
@@ -87,9 +84,6 @@ CREATE FUNCTION vector_cmp(vector, vector) RETURNS int4
|
||||
CREATE FUNCTION vector_l2_squared_distance(vector, vector) RETURNS float8
|
||||
AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
|
||||
|
||||
CREATE FUNCTION float4_l2_squared_distance(float4[], float4[]) RETURNS float8
|
||||
AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
|
||||
|
||||
CREATE FUNCTION vector_negative_inner_product(vector, vector) RETURNS float8
|
||||
AS 'MODULE_PATHNAME' LANGUAGE C IMMUTABLE STRICT PARALLEL SAFE;
|
||||
|
||||
@@ -170,11 +164,6 @@ CREATE OPERATOR <-> (
|
||||
COMMUTATOR = '<->'
|
||||
);
|
||||
|
||||
CREATE OPERATOR <-> (
|
||||
LEFTARG = float4[], RIGHTARG = float4[], PROCEDURE = l2_distance,
|
||||
COMMUTATOR = '<->'
|
||||
);
|
||||
|
||||
CREATE OPERATOR <#> (
|
||||
LEFTARG = vector, RIGHTARG = vector, PROCEDURE = vector_negative_inner_product,
|
||||
COMMUTATOR = '<#>'
|
||||
@@ -291,11 +280,6 @@ CREATE OPERATOR CLASS vector_l2_ops
|
||||
OPERATOR 1 <-> (vector, vector) FOR ORDER BY float_ops,
|
||||
FUNCTION 1 vector_l2_squared_distance(vector, vector);
|
||||
|
||||
CREATE OPERATOR CLASS float4_l2_ops
|
||||
FOR TYPE float4[] USING hnsw AS
|
||||
OPERATOR 1 <-> (float4[], float4[]) FOR ORDER BY float_ops,
|
||||
FUNCTION 1 float4_l2_squared_distance(float4[], float4[]);
|
||||
|
||||
CREATE OPERATOR CLASS vector_ip_ops
|
||||
FOR TYPE vector USING hnsw AS
|
||||
OPERATOR 1 <#> (vector, vector) FOR ORDER BY float_ops,
|
||||
|
||||
60
src/float4.c
60
src/float4.c
@@ -1,60 +0,0 @@
|
||||
#include "postgres.h"
|
||||
|
||||
#include <math.h>
|
||||
|
||||
#include "utils/array.h"
|
||||
|
||||
/*
|
||||
* Get the L2 distance between vectors
|
||||
*/
|
||||
PGDLLEXPORT PG_FUNCTION_INFO_V1(float4_l2_distance);
|
||||
Datum
|
||||
float4_l2_distance(PG_FUNCTION_ARGS)
|
||||
{
|
||||
ArrayType *a = PG_GETARG_ARRAYTYPE_P(0);
|
||||
ArrayType *b = PG_GETARG_ARRAYTYPE_P(1);
|
||||
float *ax = (float *) ARR_DATA_PTR(a);
|
||||
float *bx = (float *) ARR_DATA_PTR(b);
|
||||
float distance = 0.0;
|
||||
float diff;
|
||||
|
||||
/* TODO Check rank, dimensions, and nulls */
|
||||
int dim = ARR_DIMS(a)[0];
|
||||
|
||||
/* Auto-vectorized */
|
||||
for (int i = 0; i < dim; i++)
|
||||
{
|
||||
diff = ax[i] - bx[i];
|
||||
distance += diff * diff;
|
||||
}
|
||||
|
||||
PG_RETURN_FLOAT8(sqrt((double) distance));
|
||||
}
|
||||
|
||||
/*
|
||||
* Get the L2 squared distance between vectors
|
||||
* This saves a sqrt calculation
|
||||
*/
|
||||
PGDLLEXPORT PG_FUNCTION_INFO_V1(float4_l2_squared_distance);
|
||||
Datum
|
||||
float4_l2_squared_distance(PG_FUNCTION_ARGS)
|
||||
{
|
||||
ArrayType *a = PG_GETARG_ARRAYTYPE_P(0);
|
||||
ArrayType *b = PG_GETARG_ARRAYTYPE_P(1);
|
||||
float *ax = (float *) ARR_DATA_PTR(a);
|
||||
float *bx = (float *) ARR_DATA_PTR(b);
|
||||
float distance = 0.0;
|
||||
float diff;
|
||||
|
||||
/* TODO Check rank, dimensions, and nulls */
|
||||
int dim = ARR_DIMS(a)[0];
|
||||
|
||||
/* Auto-vectorized */
|
||||
for (int i = 0; i < dim; i++)
|
||||
{
|
||||
diff = ax[i] - bx[i];
|
||||
distance += diff * diff;
|
||||
}
|
||||
|
||||
PG_RETURN_FLOAT8((double) distance);
|
||||
}
|
||||
@@ -33,12 +33,6 @@ HnswInit(void)
|
||||
HNSW_DEFAULT_EF_CONSTRUCTION, HNSW_MIN_EF_CONSTRUCTION, HNSW_MAX_EF_CONSTRUCTION
|
||||
#if PG_VERSION_NUM >= 130000
|
||||
,AccessExclusiveLock
|
||||
#endif
|
||||
);
|
||||
add_int_reloption(hnsw_relopt_kind, "dimensions", "Number of dimensions",
|
||||
HNSW_DEFAULT_DIMENSIONS, HNSW_MIN_DIMENSIONS, HNSW_MAX_DIMENSIONS
|
||||
#if PG_VERSION_NUM >= 130000
|
||||
,AccessExclusiveLock
|
||||
#endif
|
||||
);
|
||||
|
||||
@@ -131,7 +125,6 @@ hnswoptions(Datum reloptions, bool validate)
|
||||
static const relopt_parse_elt tab[] = {
|
||||
{"m", RELOPT_TYPE_INT, offsetof(HnswOptions, m)},
|
||||
{"ef_construction", RELOPT_TYPE_INT, offsetof(HnswOptions, efConstruction)},
|
||||
{"dimensions", RELOPT_TYPE_INT, offsetof(HnswOptions, dimensions)},
|
||||
};
|
||||
|
||||
#if PG_VERSION_NUM >= 130000
|
||||
|
||||
14
src/hnsw.h
14
src/hnsw.h
@@ -42,9 +42,6 @@
|
||||
#define HNSW_DEFAULT_EF_SEARCH 40
|
||||
#define HNSW_MIN_EF_SEARCH 1
|
||||
#define HNSW_MAX_EF_SEARCH 1000
|
||||
#define HNSW_DEFAULT_DIMENSIONS -1
|
||||
#define HNSW_MIN_DIMENSIONS 1
|
||||
#define HNSW_MAX_DIMENSIONS HNSW_MAX_DIM
|
||||
|
||||
/* Tuple types */
|
||||
#define HNSW_ELEMENT_TUPLE_TYPE 1
|
||||
@@ -60,7 +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))
|
||||
@@ -99,13 +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;
|
||||
@@ -134,7 +130,6 @@ typedef struct HnswOptions
|
||||
int32 vl_len_; /* varlena header (do not touch directly!) */
|
||||
int m; /* number of connections */
|
||||
int efConstruction; /* size of dynamic candidate list */
|
||||
int dimensions;
|
||||
} HnswOptions;
|
||||
|
||||
typedef struct HnswBuildState
|
||||
@@ -205,7 +200,7 @@ typedef struct HnswElementTupleData
|
||||
ItemPointerData heaptids[HNSW_HEAPTIDS];
|
||||
ItemPointerData neighbortid;
|
||||
uint16 unused2;
|
||||
char value[FLEXIBLE_ARRAY_MEMBER];
|
||||
Vector vec;
|
||||
} HnswElementTupleData;
|
||||
|
||||
typedef HnswElementTupleData * HnswElementTuple;
|
||||
@@ -263,8 +258,7 @@ typedef struct HnswVacuumState
|
||||
/* Methods */
|
||||
int HnswGetM(Relation index);
|
||||
int HnswGetEfConstruction(Relation index);
|
||||
int HnswGetDimensions(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);
|
||||
|
||||
@@ -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);
|
||||
}
|
||||
|
||||
/*
|
||||
@@ -400,7 +398,6 @@ HnswGetMaxInMemoryElements(int m, double ml, int dimensions)
|
||||
elementSize += sizeof(HnswNeighborArray) * (avgLevel + 1);
|
||||
elementSize += sizeof(HnswCandidate) * (m * (avgLevel + 2));
|
||||
elementSize += sizeof(ItemPointerData);
|
||||
/* TODO Handle non-vector types */
|
||||
elementSize += VECTOR_SIZE(dimensions);
|
||||
return (maintenance_work_mem * 1024L) / elementSize;
|
||||
}
|
||||
@@ -418,10 +415,7 @@ InitBuildState(HnswBuildState * buildstate, Relation heap, Relation index, Index
|
||||
|
||||
buildstate->m = HnswGetM(index);
|
||||
buildstate->efConstruction = HnswGetEfConstruction(index);
|
||||
buildstate->dimensions = HnswGetDimensions(index);
|
||||
|
||||
if (buildstate->dimensions < 0)
|
||||
buildstate->dimensions = TupleDescAttr(index->rd_att, 0)->atttypmod;
|
||||
buildstate->dimensions = TupleDescAttr(index->rd_att, 0)->atttypmod;
|
||||
|
||||
/* Require column to have dimensions to be indexed */
|
||||
if (buildstate->dimensions < 0)
|
||||
|
||||
@@ -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));
|
||||
@@ -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;
|
||||
|
||||
@@ -521,7 +522,7 @@ HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_ti
|
||||
|
||||
/* Create an element */
|
||||
element = HnswInitElement(heap_tid, m, HnswGetMl(m), HnswGetMaxLevel(m));
|
||||
element->value = value;
|
||||
element->vec = DatumGetVector(value);
|
||||
|
||||
/* Prevent concurrent inserts when likely updating entry point */
|
||||
if (entryPoint == NULL || element->level > entryPoint->level)
|
||||
|
||||
@@ -4,7 +4,6 @@
|
||||
|
||||
#include "hnsw.h"
|
||||
#include "storage/bufmgr.h"
|
||||
#include "utils/datum.h"
|
||||
#include "vector.h"
|
||||
|
||||
/*
|
||||
@@ -35,30 +34,16 @@ HnswGetEfConstruction(Relation index)
|
||||
return HNSW_DEFAULT_EF_CONSTRUCTION;
|
||||
}
|
||||
|
||||
/*
|
||||
* Get the number of dimensions in the index
|
||||
*/
|
||||
int
|
||||
HnswGetDimensions(Relation index)
|
||||
{
|
||||
HnswOptions *opts = (HnswOptions *) index->rd_options;
|
||||
|
||||
if (opts)
|
||||
return opts->dimensions;
|
||||
|
||||
return HNSW_DEFAULT_DIMENSIONS;
|
||||
}
|
||||
|
||||
/*
|
||||
* 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);
|
||||
}
|
||||
|
||||
/*
|
||||
@@ -158,17 +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
|
||||
*/
|
||||
@@ -200,10 +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);
|
||||
}
|
||||
|
||||
@@ -230,7 +205,7 @@ HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno)
|
||||
element->blkno = blkno;
|
||||
element->offno = offno;
|
||||
element->neighbors = NULL;
|
||||
element->loaded = false;
|
||||
element->vec = NULL;
|
||||
return element;
|
||||
}
|
||||
|
||||
@@ -340,7 +315,7 @@ 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));
|
||||
}
|
||||
|
||||
/*
|
||||
@@ -463,10 +438,8 @@ 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));
|
||||
}
|
||||
}
|
||||
|
||||
@@ -494,7 +467,7 @@ 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);
|
||||
}
|
||||
@@ -505,7 +478,7 @@ 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)));
|
||||
}
|
||||
|
||||
/*
|
||||
@@ -740,7 +713,7 @@ 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)));
|
||||
}
|
||||
|
||||
/*
|
||||
@@ -823,7 +796,7 @@ 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 */
|
||||
@@ -898,13 +871,13 @@ 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 = ¤tNeighbors->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);
|
||||
@@ -986,7 +959,7 @@ 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 */
|
||||
|
||||
@@ -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++)
|
||||
@@ -485,7 +485,6 @@ MarkDeleted(HnswVacuumState * vacuumstate)
|
||||
HnswNeighborTuple ntup;
|
||||
Size etupSize;
|
||||
Size ntupSize;
|
||||
Datum value;
|
||||
Buffer nbuf;
|
||||
Page npage;
|
||||
BlockNumber neighborPage;
|
||||
@@ -509,11 +508,8 @@ 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 +532,7 @@ 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++)
|
||||
|
||||
@@ -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));
|
||||
|
||||
|
||||
@@ -40,6 +40,9 @@
|
||||
#define IVFFLAT_METAPAGE_BLKNO 0
|
||||
#define IVFFLAT_HEAD_BLKNO 1 /* first list page */
|
||||
|
||||
/* Must correspond to page numbers since page lock is used */
|
||||
#define IVFFLAT_SCAN_LOCK 0
|
||||
|
||||
/* IVFFlat parameters */
|
||||
#define IVFFLAT_DEFAULT_LISTS 100
|
||||
#define IVFFLAT_MIN_LISTS 1
|
||||
@@ -52,7 +55,7 @@
|
||||
#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 +232,7 @@ typedef struct IvfflatListData
|
||||
{
|
||||
BlockNumber startPage;
|
||||
BlockNumber insertPage;
|
||||
char center[FLEXIBLE_ARRAY_MEMBER];
|
||||
Vector center;
|
||||
} IvfflatListData;
|
||||
|
||||
typedef IvfflatListData * IvfflatList;
|
||||
@@ -246,6 +249,7 @@ typedef struct IvfflatScanOpaqueData
|
||||
int probes;
|
||||
int dimensions;
|
||||
bool first;
|
||||
bool hasLock;
|
||||
Buffer buf;
|
||||
ItemPointerData heaptid;
|
||||
|
||||
@@ -277,7 +281,7 @@ 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);
|
||||
|
||||
@@ -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);
|
||||
}
|
||||
|
||||
/*
|
||||
|
||||
@@ -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.");
|
||||
}
|
||||
|
||||
@@ -9,6 +9,7 @@
|
||||
#include "miscadmin.h"
|
||||
#include "pgstat.h"
|
||||
#include "storage/bufmgr.h"
|
||||
#include "storage/lmgr.h"
|
||||
|
||||
/*
|
||||
* Compare list distances
|
||||
@@ -263,6 +264,7 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
|
||||
so = (IvfflatScanOpaque) palloc(offsetof(IvfflatScanOpaqueData, lists) + probes * sizeof(IvfflatScanList));
|
||||
so->buf = InvalidBuffer;
|
||||
so->first = true;
|
||||
so->hasLock = false;
|
||||
ItemPointerSetInvalid(&so->heaptid);
|
||||
so->probes = probes;
|
||||
so->dimensions = dimensions;
|
||||
@@ -347,6 +349,13 @@ ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
|
||||
if (scan->orderByData == NULL)
|
||||
elog(ERROR, "cannot scan ivfflat index without order");
|
||||
|
||||
/* Get a shared lock for non-MVCC snapshots */
|
||||
if (!so->hasLock && !IsMVCCSnapshot(scan->xs_snapshot))
|
||||
{
|
||||
so->hasLock = true;
|
||||
LockPage(scan->indexRelation, IVFFLAT_SCAN_LOCK, ShareLock);
|
||||
}
|
||||
|
||||
if (scan->orderByData->sk_flags & SK_ISNULL)
|
||||
value = PointerGetDatum(InitVector(so->dimensions));
|
||||
else
|
||||
@@ -422,6 +431,10 @@ ivfflatendscan(IndexScanDesc scan)
|
||||
if (BufferIsValid(so->buf))
|
||||
ReleaseBuffer(so->buf);
|
||||
|
||||
/* Release lock */
|
||||
if (so->hasLock)
|
||||
UnlockPage(scan->indexRelation, IVFFLAT_SCAN_LOCK, ShareLock);
|
||||
|
||||
pairingheap_free(so->listQueue);
|
||||
tuplesort_end(so->sortstate);
|
||||
|
||||
|
||||
@@ -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);
|
||||
}
|
||||
|
||||
/*
|
||||
|
||||
@@ -3,6 +3,7 @@
|
||||
#include "commands/vacuum.h"
|
||||
#include "ivfflat.h"
|
||||
#include "storage/bufmgr.h"
|
||||
#include "storage/lmgr.h"
|
||||
|
||||
/*
|
||||
* Bulk delete tuples from the index
|
||||
@@ -65,14 +66,10 @@ ivfflatbulkdelete(IndexVacuumInfo *info, IndexBulkDeleteResult *stats,
|
||||
|
||||
vacuum_delay_point();
|
||||
|
||||
buf = ReadBufferExtended(index, MAIN_FORKNUM, searchPage, RBM_NORMAL, bas);
|
||||
/* Ensure no in-flight index scans for non-MVCC snapshots */
|
||||
LockPage(index, IVFFLAT_SCAN_LOCK, ExclusiveLock);
|
||||
|
||||
/*
|
||||
* ambulkdelete cannot delete entries from pages that are
|
||||
* pinned by other backends
|
||||
*
|
||||
* https://www.postgresql.org/docs/current/index-locking.html
|
||||
*/
|
||||
buf = ReadBufferExtended(index, MAIN_FORKNUM, searchPage, RBM_NORMAL, bas);
|
||||
LockBufferForCleanup(buf);
|
||||
|
||||
state = GenericXLogStart(index);
|
||||
@@ -114,6 +111,8 @@ ivfflatbulkdelete(IndexVacuumInfo *info, IndexBulkDeleteResult *stats,
|
||||
GenericXLogAbort(state);
|
||||
|
||||
UnlockReleaseBuffer(buf);
|
||||
|
||||
UnlockPage(index, IVFFLAT_SCAN_LOCK, ExclusiveLock);
|
||||
}
|
||||
|
||||
/*
|
||||
|
||||
@@ -54,21 +54,21 @@ SELECT vector_norm('[3e37,4e37]')::real;
|
||||
5e+37
|
||||
(1 row)
|
||||
|
||||
SELECT l2_distance('[0,0]'::vector, '[3,4]');
|
||||
SELECT l2_distance('[0,0]', '[3,4]');
|
||||
l2_distance
|
||||
-------------
|
||||
5
|
||||
(1 row)
|
||||
|
||||
SELECT l2_distance('[0,0]'::vector, '[0,1]');
|
||||
SELECT l2_distance('[0,0]', '[0,1]');
|
||||
l2_distance
|
||||
-------------
|
||||
1
|
||||
(1 row)
|
||||
|
||||
SELECT l2_distance('[1,2]'::vector, '[3]');
|
||||
SELECT l2_distance('[1,2]', '[3]');
|
||||
ERROR: different vector dimensions 2 and 1
|
||||
SELECT l2_distance('[3e38]'::vector, '[-3e38]');
|
||||
SELECT l2_distance('[3e38]', '[-3e38]');
|
||||
l2_distance
|
||||
-------------
|
||||
Infinity
|
||||
|
||||
@@ -13,10 +13,10 @@ SELECT vector_norm('[3,4]');
|
||||
SELECT vector_norm('[0,1]');
|
||||
SELECT vector_norm('[3e37,4e37]')::real;
|
||||
|
||||
SELECT l2_distance('[0,0]'::vector, '[3,4]');
|
||||
SELECT l2_distance('[0,0]'::vector, '[0,1]');
|
||||
SELECT l2_distance('[1,2]'::vector, '[3]');
|
||||
SELECT l2_distance('[3e38]'::vector, '[-3e38]');
|
||||
SELECT l2_distance('[0,0]', '[3,4]');
|
||||
SELECT l2_distance('[0,0]', '[0,1]');
|
||||
SELECT l2_distance('[1,2]', '[3]');
|
||||
SELECT l2_distance('[3e38]', '[-3e38]');
|
||||
|
||||
SELECT inner_product('[1,2]', '[3,4]');
|
||||
SELECT inner_product('[1,2]', '[3]');
|
||||
|
||||
@@ -1,93 +0,0 @@
|
||||
use strict;
|
||||
use warnings;
|
||||
use PostgresNode;
|
||||
use TestLib;
|
||||
use Test::More;
|
||||
|
||||
my $node;
|
||||
my @queries = ();
|
||||
my @expected;
|
||||
my $limit = 20;
|
||||
|
||||
sub test_recall
|
||||
{
|
||||
my ($min, $operator) = @_;
|
||||
my $correct = 0;
|
||||
my $total = 0;
|
||||
|
||||
my $explain = $node->safe_psql("postgres", qq(
|
||||
SET enable_seqscan = off;
|
||||
EXPLAIN ANALYZE SELECT i FROM tst ORDER BY v $operator '$queries[0]' LIMIT $limit;
|
||||
));
|
||||
like($explain, qr/Index Scan/);
|
||||
|
||||
for my $i (0 .. $#queries)
|
||||
{
|
||||
my $actual = $node->safe_psql("postgres", qq(
|
||||
SET enable_seqscan = off;
|
||||
SELECT i FROM tst ORDER BY v $operator '$queries[$i]' LIMIT $limit;
|
||||
));
|
||||
my @actual_ids = split("\n", $actual);
|
||||
my %actual_set = map { $_ => 1 } @actual_ids;
|
||||
|
||||
my @expected_ids = split("\n", $expected[$i]);
|
||||
|
||||
foreach (@expected_ids)
|
||||
{
|
||||
if (exists($actual_set{$_}))
|
||||
{
|
||||
$correct++;
|
||||
}
|
||||
$total++;
|
||||
}
|
||||
}
|
||||
|
||||
cmp_ok($correct / $total, ">=", $min, $operator);
|
||||
}
|
||||
|
||||
# Initialize node
|
||||
$node = get_new_node('node');
|
||||
$node->init;
|
||||
$node->start;
|
||||
|
||||
# Create table
|
||||
$node->safe_psql("postgres", "CREATE EXTENSION vector;");
|
||||
$node->safe_psql("postgres", "CREATE TABLE tst (i int4, v float4[3]);");
|
||||
$node->safe_psql("postgres",
|
||||
"INSERT INTO tst SELECT i, ARRAY[random(), random(), random()] FROM generate_series(1, 10000) i;"
|
||||
);
|
||||
|
||||
# Generate queries
|
||||
for (1 .. 20)
|
||||
{
|
||||
my $r1 = rand();
|
||||
my $r2 = rand();
|
||||
my $r3 = rand();
|
||||
push(@queries, "{$r1,$r2,$r3}");
|
||||
}
|
||||
|
||||
# Check each index type
|
||||
my @operators = ("<->");
|
||||
my @opclasses = ("float4_l2_ops");
|
||||
|
||||
for my $i (0 .. $#operators)
|
||||
{
|
||||
my $operator = $operators[$i];
|
||||
my $opclass = $opclasses[$i];
|
||||
|
||||
# Get exact results
|
||||
@expected = ();
|
||||
foreach (@queries)
|
||||
{
|
||||
my $res = $node->safe_psql("postgres", "SELECT i FROM tst ORDER BY v $operator '$_' LIMIT $limit;");
|
||||
push(@expected, $res);
|
||||
}
|
||||
|
||||
# Add index
|
||||
$node->safe_psql("postgres", "CREATE INDEX ON tst USING hnsw (v $opclass) WITH (dimensions = 3);");
|
||||
|
||||
my $min = $operator eq "<#>" ? 0.80 : 0.99;
|
||||
test_recall($min, $operator);
|
||||
}
|
||||
|
||||
done_testing();
|
||||
Reference in New Issue
Block a user