Compare commits

..

1 Commits

33 changed files with 943 additions and 1470 deletions

View File

@@ -20,6 +20,8 @@ jobs:
os: ubuntu-20.04
- postgres: 12
os: ubuntu-20.04
- postgres: 11
os: ubuntu-20.04
steps:
- uses: actions/checkout@v4
- uses: ankane/setup-postgres@v1
@@ -57,10 +59,10 @@ jobs:
- run: |
brew install cpanm
cpanm --notest IPC::Run
wget -q https://github.com/postgres/postgres/archive/refs/tags/REL_14_10.tar.gz
tar xf REL_14_10.tar.gz
- run: make prove_installcheck PROVE_FLAGS="-I ./postgres-REL_14_10/src/test/perl" PERL5LIB="/Users/runner/perl5/lib/perl5"
- run: make clean && /usr/local/opt/llvm@15/bin/scan-build --status-bugs make PG_CFLAGS="-DUSE_ASSERT_CHECKING"
wget -q https://github.com/postgres/postgres/archive/refs/tags/REL_14_5.tar.gz
tar xf REL_14_5.tar.gz
- run: make prove_installcheck PROVE_FLAGS="-I ./postgres-REL_14_5/src/test/perl" PERL5LIB="/Users/runner/perl5/lib/perl5"
- run: make clean && /usr/local/opt/llvm@15/bin/scan-build --status-bugs make
windows:
runs-on: windows-latest
if: ${{ !startsWith(github.ref_name, 'mac') }}
@@ -81,10 +83,10 @@ jobs:
if: ${{ !startsWith(github.ref_name, 'mac') && !startsWith(github.ref_name, 'windows') }}
runs-on: ubuntu-latest
container:
image: debian:12
image: debian:11
options: --platform linux/386
steps:
- run: apt-get update && apt-get install -y build-essential git libipc-run-perl postgresql-15 postgresql-server-dev-15 sudo
- run: apt-get update && apt-get install -y build-essential git libipc-run-perl postgresql-13 postgresql-server-dev-13 sudo
- run: service postgresql start
- run: |
git clone https://github.com/${{ github.repository }}.git pgvector

View File

@@ -1,22 +1,11 @@
## 0.6.1 (unreleased)
## 0.5.2 (unreleased)
- Fixed error with `ANALYZE` and vectors with different dimensions
## 0.6.0 (2024-01-29)
If upgrading with Postgres 12 or Docker, see [these notes](https://github.com/pgvector/pgvector#060).
- Added support for parallel index builds for HNSW
- Added validation for GUC parameters
- Changed storage for vector from `extended` to `external`
- Improved performance of HNSW
- Added support for on-disk parallel index builds for HNSW
- Reduced memory usage for HNSW index builds
- Reduced WAL generation for HNSW index builds
- Fixed error with logical replication
- Fixed `invalid memory alloc request size` error with HNSW index builds
- Moved Docker image to `pgvector` org
- Added Docker tags for each supported version of Postgres
- Dropped support for Postgres 11
- Fixed `invalid memory alloc request size` error with HNSW index build
## 0.5.1 (2023-10-10)
@@ -64,7 +53,7 @@ If upgrading with Postgres 12 or Docker, see [these notes](https://github.com/pg
## 0.4.0 (2023-01-11)
If upgrading with Postgres < 13, see [this note](https://github.com/pgvector/pgvector/blob/v0.4.0/README.md#040).
If upgrading with Postgres < 13, see [this note](https://github.com/pgvector/pgvector#040).
- Changed text representation for vector elements to match `real`
- Changed storage for vector from `plain` to `extended`
@@ -81,7 +70,7 @@ If upgrading with Postgres < 13, see [this note](https://github.com/pgvector/pgv
## 0.3.1 (2022-11-02)
If upgrading from 0.2.7 or 0.3.0, [recreate](https://github.com/pgvector/pgvector/blob/v0.3.1/README.md#031) all `ivfflat` indexes after upgrading to ensure all data is indexed.
If upgrading from 0.2.7 or 0.3.0, [recreate](https://github.com/pgvector/pgvector#031) all `ivfflat` indexes after upgrading to ensure all data is indexed.
- Fixed issue with inserts silently corrupting `ivfflat` indexes (introduced in 0.2.7)
- Fixed segmentation fault with index creation when lists > 6500

View File

@@ -1,4 +1,4 @@
ARG PG_MAJOR=16
ARG PG_MAJOR=15
FROM postgres:$PG_MAJOR
ARG PG_MAJOR

View File

@@ -2,7 +2,7 @@
"name": "vector",
"abstract": "Open-source vector similarity search for Postgres",
"description": "Supports L2 distance, inner product, and cosine distance",
"version": "0.6.0",
"version": "0.5.1",
"maintainer": [
"Andrew Kane <andrew@ankane.org>"
],
@@ -12,7 +12,7 @@
"prereqs": {
"runtime": {
"requires": {
"PostgreSQL": "12.0.0"
"PostgreSQL": "11.0.0"
}
}
},
@@ -20,7 +20,7 @@
"vector": {
"file": "sql/vector.sql",
"docfile": "README.md",
"version": "0.6.0",
"version": "0.5.1",
"abstract": "Open-source vector similarity search for Postgres"
}
},

View File

@@ -1,5 +1,5 @@
EXTENSION = vector
EXTVERSION = 0.6.0
EXTVERSION = 0.5.1
MODULE_big = vector
DATA = $(wildcard sql/*--*.sql)
@@ -65,17 +65,13 @@ dist:
mkdir -p dist
git archive --format zip --prefix=$(EXTENSION)-$(EXTVERSION)/ --output dist/$(EXTENSION)-$(EXTVERSION).zip master
# for Docker
PG_MAJOR ?= 16
.PHONY: docker
docker:
docker build --pull --no-cache --build-arg PG_MAJOR=$(PG_MAJOR) -t pgvector/pgvector:pg$(PG_MAJOR) .
docker build --build-arg PG_MAJOR=$(PG_MAJOR) -t pgvector/pgvector:$(EXTVERSION)-pg$(PG_MAJOR) .
docker build --pull --no-cache --platform linux/amd64 -t ankane/pgvector:latest .
.PHONY: docker-release
docker-release:
docker buildx build --push --pull --no-cache --platform linux/amd64,linux/arm64 --build-arg PG_MAJOR=$(PG_MAJOR) -t pgvector/pgvector:pg$(PG_MAJOR) .
docker buildx build --push --platform linux/amd64,linux/arm64 --build-arg PG_MAJOR=$(PG_MAJOR) -t pgvector/pgvector:$(EXTVERSION)-pg$(PG_MAJOR) .
docker buildx build --push --pull --no-cache --platform linux/amd64,linux/arm64 -t ankane/pgvector:latest .
docker buildx build --push --platform linux/amd64,linux/arm64 -t ankane/pgvector:v$(EXTVERSION) .

View File

@@ -1,5 +1,5 @@
EXTENSION = vector
EXTVERSION = 0.6.0
EXTVERSION = 0.5.1
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

105
README.md
View File

@@ -14,13 +14,11 @@ Plus [ACID](https://en.wikipedia.org/wiki/ACID) compliance, point-in-time recove
## Installation
### Linux and Mac
Compile and install the extension (supports Postgres 12+)
Compile and install the extension (supports Postgres 11+)
```sh
cd /tmp
git clone --branch v0.6.0 https://github.com/pgvector/pgvector.git
git clone --branch v0.5.1 https://github.com/pgvector/pgvector.git
cd pgvector
make
make install # may need sudo
@@ -30,28 +28,6 @@ See the [installation notes](#installation-notes) if you run into issues
You can also install it with [Docker](#docker), [Homebrew](#homebrew), [PGXN](#pgxn), [APT](#apt), [Yum](#yum), or [conda-forge](#conda-forge), and it comes preinstalled with [Postgres.app](#postgresapp) and many [hosted providers](#hosted-postgres). There are also instructions for [GitHub Actions](https://github.com/pgvector/setup-pgvector).
### Windows
Ensure [C++ support in Visual Studio](https://learn.microsoft.com/en-us/cpp/build/building-on-the-command-line?view=msvc-170#download-and-install-the-tools) is installed, and run:
```cmd
call "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\Build\vcvars64.bat"
```
Note: The exact path will vary depending on your Visual Studio version and edition
Then use `nmake` to build:
```cmd
set "PGROOT=C:\Program Files\PostgreSQL\16"
git clone --branch v0.6.0 https://github.com/pgvector/pgvector.git
cd pgvector
nmake /F Makefile.win
nmake /F Makefile.win install
```
You can also install it with [Docker](#docker) or [conda-forge](#conda-forge).
## Getting Started
Enable the extension (do this once in each database where you want to use it)
@@ -264,14 +240,6 @@ HINT: Increase maintenance_work_mem to speed up builds.
Note: Do not set `maintenance_work_mem` so high that it exhausts the memory on the server
Starting with 0.6.0, you can also speed up index creation by increasing the number of parallel workers (2 by default)
```sql
SET max_parallel_maintenance_workers = 7; -- plus leader
```
For a large number of workers, you may also need to increase `max_parallel_workers` (8 by default)
### Indexing Progress
Check [indexing progress](https://www.postgresql.org/docs/current/progress-reporting.html#CREATE-INDEX-PROGRESS-REPORTING) with Postgres 12+
@@ -430,15 +398,6 @@ To speed up queries with an IVFFlat index, increase the number of inverted lists
CREATE INDEX ON items USING ivfflat (embedding vector_l2_ops) WITH (lists = 1000);
```
## Vacuuming
Vacuuming can take a while for HNSW indexes. Speed it up by reindexing first.
```sql
REINDEX INDEX CONCURRENTLY index_name;
VACUUM table_name;
```
## Languages
Use pgvector from any language with a Postgres client. You can even generate and store vectors in one language and query them in another.
@@ -656,24 +615,44 @@ Note: Replace `16` with your Postgres server version
If compilation fails and the output includes `warning: no such sysroot directory` on Mac, reinstall Xcode Command Line Tools.
### Windows
Support for Windows is currently experimental. Ensure [C++ support in Visual Studio](https://learn.microsoft.com/en-us/cpp/build/building-on-the-command-line?view=msvc-170#download-and-install-the-tools) is installed, and run:
```cmd
call "C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Auxiliary\Build\vcvars64.bat"
```
Note: The exact path will vary depending on your Visual Studio version and edition
Then use `nmake` to build:
```cmd
set "PGROOT=C:\Program Files\PostgreSQL\16"
git clone --branch v0.5.1 https://github.com/pgvector/pgvector.git
cd pgvector
nmake /F Makefile.win
nmake /F Makefile.win install
```
## Additional Installation Methods
### Docker
Get the [Docker image](https://hub.docker.com/r/pgvector/pgvector) with:
Get the [Docker image](https://hub.docker.com/r/ankane/pgvector) with:
```sh
docker pull pgvector/pgvector:pg16
docker pull ankane/pgvector
```
This adds pgvector to the [Postgres image](https://hub.docker.com/_/postgres) (replace `16` with your Postgres server version, and run it the same way).
This adds pgvector to the [Postgres image](https://hub.docker.com/_/postgres) (run it the same way).
You can also build the image manually:
```sh
git clone --branch v0.6.0 https://github.com/pgvector/pgvector.git
git clone --branch v0.5.1 https://github.com/pgvector/pgvector.git
cd pgvector
docker build --build-arg PG_MAJOR=16 -t myuser/pgvector .
docker build --build-arg PG_MAJOR=15 -t myuser/pgvector .
```
### Homebrew
@@ -750,32 +729,28 @@ SELECT extversion FROM pg_extension WHERE extname = 'vector';
## Upgrade Notes
### 0.6.0
### 0.4.0
#### Postgres 12
If upgrading with Postgres 12, remove this line from `sql/vector--0.5.1--0.6.0.sql`:
If upgrading with Postgres < 13, remove this line from `sql/vector--0.3.2--0.4.0.sql`:
```sql
ALTER TYPE vector SET (STORAGE = external);
ALTER TYPE vector SET (STORAGE = extended);
```
Then run `make install` and `ALTER EXTENSION vector UPDATE;`.
#### Docker
### 0.3.1
The Docker image is now published in the `pgvector` org, and there are tags for each supported version of Postgres (rather than a `latest` tag).
If upgrading from 0.2.7 or 0.3.0, recreate all `ivfflat` indexes after upgrading to ensure all data is indexed.
```sh
docker pull pgvector/pgvector:pg16
# or
docker pull pgvector/pgvector:0.6.0-pg16
```
```sql
-- Postgres 12+
REINDEX INDEX CONCURRENTLY index_name;
Also, if youve increased `maintenance_work_mem`, make sure `--shm-size` is at least that size to avoid an error with parallel HNSW index builds.
```sh
docker run --shm-size=1g ...
-- Postgres < 12
CREATE INDEX CONCURRENTLY temp_name ON table USING ivfflat (column opclass);
DROP INDEX CONCURRENTLY index_name;
ALTER INDEX temp_name RENAME TO index_name;
```
## Thanks
@@ -822,7 +797,7 @@ To run single tests:
```sh
make installcheck REGRESS=functions # regression test
make prove_installcheck PROVE_TESTS=test/t/001_ivfflat_wal.pl # TAP test
make prove_installcheck PROVE_TESTS=test/t/001_wal.pl # TAP test
```
To enable benchmarking:

View File

@@ -1,5 +0,0 @@
-- complain if script is sourced in psql, rather than via CREATE EXTENSION
\echo Use "ALTER EXTENSION vector UPDATE TO '0.6.0'" to load this file. \quit
-- remove this single line for Postgres < 13
ALTER TYPE vector SET (STORAGE = external);

View File

@@ -26,7 +26,7 @@ CREATE TYPE vector (
TYPMOD_IN = vector_typmod_in,
RECEIVE = vector_recv,
SEND = vector_send,
STORAGE = external
STORAGE = extended
);
-- functions

View File

@@ -4,57 +4,25 @@
#include <math.h>
#include "access/amapi.h"
#include "access/reloptions.h"
#include "commands/progress.h"
#include "commands/vacuum.h"
#include "hnsw.h"
#include "utils/guc.h"
#include "utils/selfuncs.h"
#if PG_VERSION_NUM < 150000
#define MarkGUCPrefixReserved(x) EmitWarningsOnPlaceholders(x)
#if PG_VERSION_NUM >= 120000
#include "commands/progress.h"
#endif
int hnsw_ef_search;
int hnsw_lock_tranche_id;
bool hnsw_enable_parallel_build;
static relopt_kind hnsw_relopt_kind;
/*
* Assign a tranche ID for our LWLocks. This only needs to be done by one
* backend, as the tranche ID is remembered in shared memory.
*
* This shared memory area is very small, so we just allocate it from the
* "slop" that PostgreSQL reserves for small allocations like this. If
* this grows bigger, we should use a shmem_request_hook and
* RequestAddinShmemSpace() to pre-reserve space for this.
*/
static void
HnswInitLockTranche(void)
{
int *tranche_ids;
bool found;
LWLockAcquire(AddinShmemInitLock, LW_EXCLUSIVE);
tranche_ids = ShmemInitStruct("hnsw LWLock ids",
sizeof(int) * 1,
&found);
if (!found)
tranche_ids[0] = LWLockNewTrancheId();
hnsw_lock_tranche_id = tranche_ids[0];
LWLockRelease(AddinShmemInitLock);
/* Per-backend registration of the tranche ID */
LWLockRegisterTranche(hnsw_lock_tranche_id, "HnswBuild");
}
/*
* Initialize index options and variables
*/
void
HnswInit(void)
{
HnswInitLockTranche();
hnsw_relopt_kind = add_reloption_kind();
add_int_reloption(hnsw_relopt_kind, "m", "Max number of connections",
HNSW_DEFAULT_M, HNSW_MIN_M, HNSW_MAX_M
@@ -73,12 +41,16 @@ HnswInit(void)
"Valid range is 1..1000.", &hnsw_ef_search,
HNSW_DEFAULT_EF_SEARCH, HNSW_MIN_EF_SEARCH, HNSW_MAX_EF_SEARCH, PGC_USERSET, 0, NULL, NULL, NULL);
MarkGUCPrefixReserved("hnsw");
/* Behind a variable for now since can be slower than building in memory */
DefineCustomBoolVariable("hnsw.enable_parallel_build", "Enables or disables building indexes in parallel",
NULL, &hnsw_enable_parallel_build,
false, PGC_USERSET, 0, NULL, NULL, NULL);
}
/*
* Get the name of index build phase
*/
#if PG_VERSION_NUM >= 120000
static char *
hnswbuildphasename(int64 phasenum)
{
@@ -92,6 +64,7 @@ hnswbuildphasename(int64 phasenum)
return NULL;
}
}
#endif
/*
* Estimate the cost of an index scan
@@ -106,6 +79,9 @@ hnswcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
int m;
int entryLevel;
Relation index;
#if PG_VERSION_NUM < 120000
List *qinfos;
#endif
/* Never use index without order */
if (path->indexorderbys == NULL)
@@ -118,6 +94,20 @@ hnswcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
return;
}
/*
* Do not use index if no limit or limit + offset > ef_search unless
* enable_seqscan = off
*/
if (root->limit_tuples < 0 || root->limit_tuples > hnsw_ef_search)
{
*indexStartupCost = 1.0e10 - 1;
*indexTotalCost = 1.0e10 - 1;
*indexSelectivity = 0;
*indexCorrelation = 0;
*indexPages = 0;
return;
}
MemSet(&costs, 0, sizeof(costs));
index = index_open(path->indexinfo->indexoid, NoLock);
@@ -131,7 +121,12 @@ hnswcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
/* Account for number of tuples (or entry level), m, and ef_search */
costs.numIndexTuples = (entryLevel + 2) * m;
#if PG_VERSION_NUM >= 120000
genericcostestimate(root, path, loop_count, &costs);
#else
qinfos = deconstruct_indexquals(path);
genericcostestimate(root, path, loop_count, qinfos, &costs);
#endif
/* Use total cost since most work happens before first tuple is returned */
*indexStartupCost = costs.indexTotalCost;
@@ -225,7 +220,9 @@ hnswhandler(PG_FUNCTION_ARGS)
amroutine->amcostestimate = hnswcostestimate;
amroutine->amoptions = hnswoptions;
amroutine->amproperty = NULL; /* TODO AMPROP_DISTANCE_ORDERABLE */
#if PG_VERSION_NUM >= 120000
amroutine->ambuildphasename = hnswbuildphasename;
#endif
amroutine->amvalidate = hnswvalidate;
#if PG_VERSION_NUM >= 140000
amroutine->amadjustmembers = NULL;

View File

@@ -3,17 +3,21 @@
#include "postgres.h"
#include "access/genam.h"
#include "access/generic_xlog.h"
#include "access/parallel.h"
#include "lib/pairingheap.h"
#include "access/reloptions.h"
#include "lib/ilist.h"
#include "nodes/execnodes.h"
#include "port.h" /* for random() */
#include "utils/relptr.h"
#include "utils/sampling.h"
#include "vector.h"
#if PG_VERSION_NUM < 110000
#error "Requires PostgreSQL 11+"
#endif
#if PG_VERSION_NUM < 120000
#error "Requires PostgreSQL 12+"
#include "access/relscan.h"
#endif
#define HNSW_MAX_DIM 2000
@@ -60,13 +64,10 @@
#define PROGRESS_HNSW_PHASE_LOAD 2
#define HNSW_MAX_SIZE (BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(HnswPageOpaqueData)) - sizeof(ItemIdData))
#define HNSW_TUPLE_ALLOC_SIZE BLCKSZ
#define HNSW_ELEMENT_TUPLE_SIZE(size) MAXALIGN(offsetof(HnswElementTupleData, data) + (size))
#define HNSW_NEIGHBOR_TUPLE_SIZE(level, m) MAXALIGN(offsetof(HnswNeighborTupleData, indextids) + ((level) + 2) * (m) * sizeof(ItemPointerData))
#define HNSW_NEIGHBOR_ARRAY_SIZE(lm) (offsetof(HnswNeighborArray, items) + sizeof(HnswCandidate) * (lm))
#define HnswPageGetOpaque(page) ((HnswPageOpaque) PageGetSpecialPointer(page))
#define HnswPageGetMeta(page) ((HnswMetaPageData *) PageGetContents(page))
@@ -95,62 +96,33 @@
/* Ensure fits on page and in uint8 */
#define HnswGetMaxLevel(m) Min(((BLCKSZ - MAXALIGN(SizeOfPageHeaderData) - MAXALIGN(sizeof(HnswPageOpaqueData)) - offsetof(HnswNeighborTupleData, indextids) - sizeof(ItemIdData)) / (sizeof(ItemPointerData)) / (m)) - 2, 255)
#define HnswGetValue(base, element) PointerGetDatum(HnswPtrAccess(base, (element)->value))
#if PG_VERSION_NUM < 140005
#define relptr_offset(rp) ((rp).relptr_off - 1)
#endif
/* Pointer macros */
#define HnswPtrAccess(base, hp) ((base) == NULL ? (hp).ptr : relptr_access(base, (hp).relptr))
#define HnswPtrStore(base, hp, value) ((base) == NULL ? (void) ((hp).ptr = (value)) : (void) relptr_store(base, (hp).relptr, value))
#define HnswPtrIsNull(base, hp) ((base) == NULL ? (hp).ptr == NULL : relptr_is_null((hp).relptr))
#define HnswPtrEqual(base, hp1, hp2) ((base) == NULL ? (hp1).ptr == (hp2).ptr : relptr_offset((hp1).relptr) == relptr_offset((hp2).relptr))
/* For code paths dedicated to each type */
#define HnswPtrPointer(hp) (hp).ptr
#define HnswPtrOffset(hp) relptr_offset((hp).relptr)
#define HnswGetNeighbors(element, lc) (AssertMacro((element)->level >= (lc)), &(element)->neighbors[lc])
/* Variables */
extern int hnsw_ef_search;
extern int hnsw_lock_tranche_id;
typedef struct HnswElementData HnswElementData;
typedef struct HnswNeighborArray HnswNeighborArray;
#define HnswPtrDeclare(type, relptrtype, ptrtype) \
relptr_declare(type, relptrtype); \
typedef union { type *ptr; relptrtype relptr; } ptrtype;
/* Pointers that can be absolute or relative */
/* Use char for DatumPtr so works with Pointer */
HnswPtrDeclare(HnswElementData, HnswElementRelptr, HnswElementPtr);
HnswPtrDeclare(HnswNeighborArray, HnswNeighborArrayRelptr, HnswNeighborArrayPtr);
HnswPtrDeclare(HnswNeighborArrayPtr, HnswNeighborsRelptr, HnswNeighborsPtr);
HnswPtrDeclare(char, DatumRelptr, DatumPtr);
extern bool hnsw_enable_parallel_build;
typedef struct HnswElementData
{
HnswElementPtr next;
slist_node next;
ItemPointerData heaptids[HNSW_HEAPTIDS];
uint8 heaptidsLength;
uint8 level;
uint8 deleted;
uint32 hash;
HnswNeighborsPtr neighbors;
struct HnswNeighborArray *neighbors;
BlockNumber blkno;
OffsetNumber offno;
OffsetNumber neighborOffno;
BlockNumber neighborPage;
DatumPtr value;
LWLock lock;
Datum value;
} HnswElementData;
typedef HnswElementData * HnswElement;
typedef struct HnswCandidate
{
HnswElementPtr element;
HnswElement element;
float distance;
bool closer;
} HnswCandidate;
@@ -159,7 +131,7 @@ typedef struct HnswNeighborArray
{
int length;
bool closerSet;
HnswCandidate items[FLEXIBLE_ARRAY_MEMBER];
HnswCandidate *items;
} HnswNeighborArray;
typedef struct HnswPairingHeapNode
@@ -178,31 +150,27 @@ typedef struct HnswOptions
typedef struct HnswGraph
{
/* Graph state */
slock_t lock;
HnswElementPtr head;
double indtuples;
/* Entry state */
LWLock entryLock;
HnswElementPtr entryPoint;
/* Allocations state */
LWLock allocatorLock;
slist_head elements;
HnswElement entryPoint;
long memoryUsed;
long memoryTotal;
/* Flushed state */
LWLock flushLock;
bool flushed;
double indtuples;
} HnswGraph;
typedef struct HnswSpool
{
Relation heap;
Relation index;
} HnswSpool;
typedef struct HnswShared
{
/* Immutable state */
Oid heaprelid;
Oid indexrelid;
bool isconcurrent;
int scantuplesortstates;
/* Worker progress */
ConditionVariable workersdonecv;
@@ -214,10 +182,16 @@ typedef struct HnswShared
int nparticipantsdone;
double reltuples;
HnswGraph graphData;
#if PG_VERSION_NUM < 120000
ParallelHeapScanDescData heapdesc; /* must come last */
#endif
} HnswShared;
#if PG_VERSION_NUM >= 120000
#define ParallelTableScanFromHnswShared(shared) \
(ParallelTableScanDesc) ((char *) (shared) + BUFFERALIGN(sizeof(HnswShared)))
#endif
typedef struct HnswLeader
{
@@ -225,15 +199,8 @@ typedef struct HnswLeader
int nparticipanttuplesorts;
HnswShared *hnswshared;
Snapshot snapshot;
char *hnswarea;
} HnswLeader;
typedef struct HnswAllocator
{
void *(*alloc) (Size size, void *state);
void *state;
} HnswAllocator;
typedef struct HnswBuildState
{
/* Info */
@@ -266,12 +233,10 @@ typedef struct HnswBuildState
/* Memory */
MemoryContext graphCtx;
MemoryContext tmpCtx;
HnswAllocator allocator;
/* Parallel builds */
HnswLeader *hnswleader;
HnswShared *hnswshared;
char *hnswarea;
} HnswBuildState;
typedef struct HnswMetaPageData
@@ -370,24 +335,23 @@ bool HnswNormValue(FmgrInfo *procinfo, Oid collation, Datum *value, Vector * re
Buffer HnswNewBuffer(Relation index, ForkNumber forkNum);
void HnswInitPage(Buffer buf, Page page);
void HnswInit(void);
List *HnswSearchLayer(char *base, 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, int m, bool inserting, HnswElement skipElement);
HnswElement HnswGetEntryPoint(Relation index);
void HnswGetMetaPageInfo(Relation index, int *m, HnswElement * entryPoint);
void *HnswAlloc(HnswAllocator * allocator, Size size);
HnswElement HnswInitElement(char *base, ItemPointer tid, int m, double ml, int maxLevel, HnswAllocator * alloc);
HnswElement HnswInitElement(ItemPointer tid, int m, double ml, int maxLevel);
HnswElement HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno);
void HnswFindElementNeighbors(char *base, HnswElement element, HnswElement entryPoint, Relation index, FmgrInfo *procinfo, Oid collation, int m, int efConstruction, bool existing);
HnswCandidate *HnswEntryCandidate(char *base, HnswElement em, Datum q, Relation rel, FmgrInfo *procinfo, Oid collation, bool loadVec);
void HnswInsertElement(HnswElement element, HnswElement entryPoint, Relation index, FmgrInfo *procinfo, Oid collation, int m, int efConstruction, bool existing);
HnswCandidate *HnswEntryCandidate(HnswElement em, Datum q, Relation rel, FmgrInfo *procinfo, Oid collation, bool loadVec);
void HnswUpdateMetaPage(Relation index, int updateEntry, HnswElement entryPoint, BlockNumber insertPage, ForkNumber forkNum, bool building);
void HnswSetNeighborTuple(char *base, HnswNeighborTuple ntup, HnswElement e, int m);
void HnswSetNeighborTuple(HnswNeighborTuple ntup, HnswElement e, int m);
void HnswAddHeapTid(HnswElement element, ItemPointer heaptid);
void HnswInitNeighbors(char *base, HnswElement element, int m, HnswAllocator * alloc);
bool HnswInsertTupleOnDisk(Relation index, Datum value, Datum *values, bool *isnull, ItemPointer heap_tid, bool building);
void HnswUpdateNeighborsOnDisk(Relation index, FmgrInfo *procinfo, Oid collation, HnswElement e, int m, bool checkExisting, bool building);
void HnswInitNeighbors(HnswElement element, int m);
bool HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, Relation heapRel, bool building);
void HnswUpdateNeighborPages(Relation index, FmgrInfo *procinfo, Oid collation, HnswElement e, int m, bool checkExisting, bool building);
void HnswLoadElementFromTuple(HnswElement element, HnswElementTuple etup, bool loadHeaptids, bool loadVec);
void HnswLoadElement(HnswElement element, float *distance, Datum *q, Relation index, FmgrInfo *procinfo, Oid collation, bool loadVec);
void HnswSetElementTuple(char *base, HnswElementTuple etup, HnswElement element);
void HnswUpdateConnection(char *base, HnswElement element, HnswCandidate * hc, int lm, int lc, int *updateIdx, Relation index, FmgrInfo *procinfo, Oid collation);
void HnswSetElementTuple(HnswElementTuple etup, HnswElement element);
void HnswUpdateConnection(HnswElement element, HnswCandidate * hc, int m, int lc, int *updateIdx, Relation index, FmgrInfo *procinfo, Oid collation);
void HnswLoadNeighbors(HnswElement element, Relation index, int m);
PGDLLEXPORT void HnswParallelBuildMain(dsm_segment *seg, shm_toc *toc);
@@ -407,16 +371,6 @@ void hnswrescan(IndexScanDesc scan, ScanKey keys, int nkeys, ScanKey orderbys,
bool hnswgettuple(IndexScanDesc scan, ScanDirection dir);
void hnswendscan(IndexScanDesc scan);
static inline HnswNeighborArray *
HnswGetNeighbors(char *base, HnswElement element, int lc)
{
HnswNeighborArrayPtr *neighborList = HnswPtrAccess(base, element->neighbors);
Assert(element->level >= lc);
return HnswPtrAccess(base, neighborList[lc]);
}
/* Hash tables */
typedef struct TidHashEntry
{
@@ -444,17 +398,4 @@ typedef struct PointerHashEntry
#define SH_DECLARE
#include "lib/simplehash.h"
typedef struct OffsetHashEntry
{
Size offset;
char status;
} OffsetHashEntry;
#define SH_PREFIX offsethash
#define SH_ELEMENT_TYPE OffsetHashEntry
#define SH_KEY_TYPE Size
#define SH_SCOPE extern
#define SH_DECLARE
#include "lib/simplehash.h"
#endif

File diff suppressed because it is too large Load Diff

View File

@@ -2,7 +2,6 @@
#include <math.h>
#include "access/generic_xlog.h"
#include "hnsw.h"
#include "storage/bufmgr.h"
#include "storage/lmgr.h"
@@ -117,7 +116,7 @@ HnswInsertAppendPage(Relation index, Buffer *nbuf, Page *npage, GenericXLogState
* Add to element and neighbor pages
*/
static void
AddElementOnDisk(Relation index, HnswElement e, int m, BlockNumber insertPage, BlockNumber *updatedInsertPage, bool building)
WriteNewElementPages(Relation index, HnswElement e, int m, BlockNumber insertPage, BlockNumber *updatedInsertPage, bool building)
{
Buffer buf;
Page page;
@@ -135,10 +134,9 @@ AddElementOnDisk(Relation index, HnswElement e, int m, BlockNumber insertPage, B
OffsetNumber freeOffno = InvalidOffsetNumber;
OffsetNumber freeNeighborOffno = InvalidOffsetNumber;
BlockNumber newInsertPage = InvalidBlockNumber;
char *base = NULL;
/* Calculate sizes */
etupSize = HNSW_ELEMENT_TUPLE_SIZE(VARSIZE_ANY(HnswPtrAccess(base, e->value)));
etupSize = HNSW_ELEMENT_TUPLE_SIZE(VARSIZE_ANY(DatumGetPointer(e->value)));
ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(e->level, m);
combinedSize = etupSize + ntupSize + sizeof(ItemIdData);
maxSize = HNSW_MAX_SIZE;
@@ -146,11 +144,11 @@ AddElementOnDisk(Relation index, HnswElement e, int m, BlockNumber insertPage, B
/* Prepare element tuple */
etup = palloc0(etupSize);
HnswSetElementTuple(base, etup, e);
HnswSetElementTuple(etup, e);
/* Prepare neighbor tuple */
ntup = palloc0(ntupSize);
HnswSetNeighborTuple(base, ntup, e, m);
HnswSetNeighborTuple(ntup, e, m);
/* Find a page (or two if needed) to insert the tuples */
for (;;)
@@ -340,14 +338,12 @@ ConnectionExists(HnswElement e, HnswNeighborTuple ntup, int startIdx, int lm)
* Update neighbors
*/
void
HnswUpdateNeighborsOnDisk(Relation index, FmgrInfo *procinfo, Oid collation, HnswElement e, int m, bool checkExisting, bool building)
HnswUpdateNeighborPages(Relation index, FmgrInfo *procinfo, Oid collation, HnswElement e, int m, bool checkExisting, bool building)
{
char *base = NULL;
for (int lc = e->level; lc >= 0; lc--)
{
int lm = HnswGetLayerM(m, lc);
HnswNeighborArray *neighbors = HnswGetNeighbors(base, e, lc);
HnswNeighborArray *neighbors = HnswGetNeighbors(e, lc);
for (int i = 0; i < neighbors->length; i++)
{
@@ -355,15 +351,16 @@ HnswUpdateNeighborsOnDisk(Relation index, FmgrInfo *procinfo, Oid collation, Hns
Buffer buf;
Page page;
GenericXLogState *state;
ItemId itemid;
HnswNeighborTuple ntup;
Size ntupSize;
int idx = -1;
int startIdx;
HnswElement neighborElement = HnswPtrAccess(base, hc->element);
OffsetNumber offno = neighborElement->neighborOffno;
OffsetNumber offno = hc->element->neighborOffno;
/* Get latest neighbors since they may have changed */
/* Do not lock yet since selecting neighbors can take time */
HnswLoadNeighbors(neighborElement, index, m);
HnswLoadNeighbors(hc->element, index, m);
/*
* Could improve performance for vacuuming by checking neighbors
@@ -373,14 +370,14 @@ HnswUpdateNeighborsOnDisk(Relation index, FmgrInfo *procinfo, Oid collation, Hns
*/
/* Select neighbors */
HnswUpdateConnection(NULL, e, hc, lm, lc, &idx, index, procinfo, collation);
HnswUpdateConnection(e, hc, lm, lc, &idx, index, procinfo, collation);
/* New element was not selected as a neighbor */
if (idx == -1)
continue;
/* Register page */
buf = ReadBuffer(index, neighborElement->neighborPage);
buf = ReadBuffer(index, hc->element->neighborPage);
LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
if (building)
{
@@ -394,10 +391,12 @@ HnswUpdateNeighborsOnDisk(Relation index, FmgrInfo *procinfo, Oid collation, Hns
}
/* Get tuple */
ntup = (HnswNeighborTuple) PageGetItem(page, PageGetItemId(page, offno));
itemid = PageGetItemId(page, offno);
ntup = (HnswNeighborTuple) PageGetItem(page, itemid);
ntupSize = ItemIdGetLength(itemid);
/* Calculate index for update */
startIdx = (neighborElement->level - lc) * m;
startIdx = (hc->element->level - lc) * m;
/* Check for existing connection */
if (checkExisting && ConnectionExists(e, ntup, startIdx, lm))
@@ -423,9 +422,13 @@ HnswUpdateNeighborsOnDisk(Relation index, FmgrInfo *procinfo, Oid collation, Hns
{
ItemPointer indextid = &ntup->indextids[idx];
/* Update neighbor on the buffer */
/* Update neighbor */
ItemPointerSet(indextid, e->blkno, e->offno);
/* Overwrite tuple */
if (!PageIndexTupleOverwrite(page, offno, (Item) ntup, ntupSize))
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
/* Commit */
if (building)
MarkBufferDirty(buf);
@@ -444,12 +447,14 @@ HnswUpdateNeighborsOnDisk(Relation index, FmgrInfo *procinfo, Oid collation, Hns
* Add a heap TID to an existing element
*/
static bool
AddDuplicateOnDisk(Relation index, HnswElement element, HnswElement dup, bool building)
HnswAddDuplicate(Relation index, HnswElement element, HnswElement dup, bool building)
{
Buffer buf;
Page page;
GenericXLogState *state;
ItemId itemid;
HnswElementTuple etup;
Size etupSize;
int i;
/* Read page */
@@ -467,7 +472,9 @@ AddDuplicateOnDisk(Relation index, HnswElement element, HnswElement dup, bool bu
}
/* Find space */
etup = (HnswElementTuple) PageGetItem(page, PageGetItemId(page, dup->offno));
itemid = PageGetItemId(page, dup->offno);
etup = (HnswElementTuple) PageGetItem(page, itemid);
etupSize = ItemIdGetLength(itemid);
for (i = 0; i < HNSW_HEAPTIDS; i++)
{
if (!ItemPointerIsValid(&etup->heaptids[i]))
@@ -483,9 +490,13 @@ AddDuplicateOnDisk(Relation index, HnswElement element, HnswElement dup, bool bu
return false;
}
/* Add heap TID, modifying the tuple on the page directly */
/* Add heap TID */
etup->heaptids[i] = element->heaptids[0];
/* Overwrite tuple */
if (!PageIndexTupleOverwrite(page, dup->offno, (Item) etup, etupSize))
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
/* Commit */
if (building)
MarkBufferDirty(buf);
@@ -500,23 +511,19 @@ AddDuplicateOnDisk(Relation index, HnswElement element, HnswElement dup, bool bu
* Find duplicate element
*/
static bool
FindDuplicateOnDisk(Relation index, HnswElement element, bool building)
HnswFindDuplicate(Relation index, HnswElement element, bool building)
{
char *base = NULL;
HnswNeighborArray *neighbors = HnswGetNeighbors(base, element, 0);
Datum value = HnswGetValue(base, element);
HnswNeighborArray *neighbors = HnswGetNeighbors(element, 0);
for (int i = 0; i < neighbors->length; i++)
{
HnswCandidate *neighbor = &neighbors->items[i];
HnswElement neighborElement = HnswPtrAccess(base, neighbor->element);
Datum neighborValue = HnswGetValue(base, neighborElement);
/* Exit early since ordered by distance */
if (!datumIsEqual(value, neighborValue, false, -1))
if (!datumIsEqual(element->value, neighbor->element->value, false, -1))
return false;
if (AddDuplicateOnDisk(index, element, neighborElement, building))
if (HnswAddDuplicate(index, element, neighbor->element, building))
return true;
}
@@ -524,26 +531,26 @@ FindDuplicateOnDisk(Relation index, HnswElement element, bool building)
}
/*
* Update graph on disk
* Write changes to disk
*/
static void
UpdateGraphOnDisk(Relation index, FmgrInfo *procinfo, Oid collation, HnswElement element, int m, int efConstruction, HnswElement entryPoint, bool building)
WriteElement(Relation index, FmgrInfo *procinfo, Oid collation, HnswElement element, int m, int efConstruction, HnswElement entryPoint, bool building)
{
BlockNumber newInsertPage = InvalidBlockNumber;
/* Look for duplicate */
if (FindDuplicateOnDisk(index, element, building))
if (HnswFindDuplicate(index, element, building))
return;
/* Add element */
AddElementOnDisk(index, element, m, GetInsertPage(index), &newInsertPage, building);
/* Write element and neighbor tuples */
WriteNewElementPages(index, element, m, GetInsertPage(index), &newInsertPage, building);
/* Update insert page if needed */
if (BlockNumberIsValid(newInsertPage))
HnswUpdateMetaPage(index, 0, NULL, newInsertPage, MAIN_FORKNUM, building);
/* Update neighbors */
HnswUpdateNeighborsOnDisk(index, procinfo, collation, element, m, false, building);
HnswUpdateNeighborPages(index, procinfo, collation, element, m, false, building);
/* Update entry point if needed */
if (entryPoint == NULL || element->level > entryPoint->level)
@@ -554,8 +561,10 @@ UpdateGraphOnDisk(Relation index, FmgrInfo *procinfo, Oid collation, HnswElement
* Insert a tuple into the index
*/
bool
HnswInsertTupleOnDisk(Relation index, Datum value, Datum *values, bool *isnull, ItemPointer heap_tid, bool building)
HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid, Relation heapRel, bool building)
{
Datum value;
FmgrInfo *normprocinfo;
HnswElement entryPoint;
HnswElement element;
int m;
@@ -563,7 +572,17 @@ HnswInsertTupleOnDisk(Relation index, Datum value, Datum *values, bool *isnull,
FmgrInfo *procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC);
Oid collation = index->rd_indcollation[0];
LOCKMODE lockmode = ShareLock;
char *base = NULL;
/* Detoast once for all calls */
value = PointerGetDatum(PG_DETOAST_DATUM(values[0]));
/* Normalize if needed */
normprocinfo = HnswOptionalProcInfo(index, HNSW_NORM_PROC);
if (normprocinfo != NULL)
{
if (!HnswNormValue(normprocinfo, collation, &value, NULL))
return false;
}
/*
* Get a shared lock. This allows vacuum to ensure no in-flight inserts
@@ -576,8 +595,8 @@ HnswInsertTupleOnDisk(Relation index, Datum value, Datum *values, bool *isnull,
HnswGetMetaPageInfo(index, &m, &entryPoint);
/* Create an element */
element = HnswInitElement(base, heap_tid, m, HnswGetMl(m), HnswGetMaxLevel(m), NULL);
HnswPtrStore(base, element->value, DatumGetPointer(value));
element = HnswInitElement(heap_tid, m, HnswGetMl(m), HnswGetMaxLevel(m));
element->value = value;
/* Prevent concurrent inserts when likely updating entry point */
if (entryPoint == NULL || element->level > entryPoint->level)
@@ -593,11 +612,11 @@ HnswInsertTupleOnDisk(Relation index, Datum value, Datum *values, bool *isnull,
entryPoint = HnswGetEntryPoint(index);
}
/* Find neighbors for element */
HnswFindElementNeighbors(base, element, entryPoint, index, procinfo, collation, m, efConstruction, false);
/* Insert element in graph */
HnswInsertElement(element, entryPoint, index, procinfo, collation, m, efConstruction, false);
/* Update graph on disk */
UpdateGraphOnDisk(index, procinfo, collation, element, m, efConstruction, entryPoint, building);
/* Write to disk */
WriteElement(index, procinfo, collation, element, m, efConstruction, entryPoint, building);
/* Release lock */
UnlockPage(index, HNSW_UPDATE_LOCK, lockmode);
@@ -605,30 +624,6 @@ HnswInsertTupleOnDisk(Relation index, Datum value, Datum *values, bool *isnull,
return true;
}
/*
* Insert a tuple into the index
*/
static void
HnswInsertTuple(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid)
{
Datum value;
FmgrInfo *normprocinfo;
Oid collation = index->rd_indcollation[0];
/* Detoast once for all calls */
value = PointerGetDatum(PG_DETOAST_DATUM(values[0]));
/* Normalize if needed */
normprocinfo = HnswOptionalProcInfo(index, HNSW_NORM_PROC);
if (normprocinfo != NULL)
{
if (!HnswNormValue(normprocinfo, collation, &value, NULL))
return;
}
HnswInsertTupleOnDisk(index, value, values, isnull, heap_tid, false);
}
/*
* Insert a tuple into the index
*/
@@ -655,7 +650,7 @@ hnswinsert(Relation index, Datum *values, bool *isnull, ItemPointer heap_tid,
oldCtx = MemoryContextSwitchTo(insertCtx);
/* Insert tuple */
HnswInsertTuple(index, values, isnull, heap_tid);
HnswInsertTuple(index, values, isnull, heap_tid, heap, false);
/* Delete memory context */
MemoryContextSwitchTo(oldCtx);

View File

@@ -21,7 +21,6 @@ GetScanItems(IndexScanDesc scan, Datum q)
List *w;
int m;
HnswElement entryPoint;
char *base = NULL;
/* Get m and entry point */
HnswGetMetaPageInfo(index, &m, &entryPoint);
@@ -29,15 +28,15 @@ GetScanItems(IndexScanDesc scan, Datum q)
if (entryPoint == NULL)
return NIL;
ep = list_make1(HnswEntryCandidate(base, entryPoint, q, index, procinfo, collation, false));
ep = list_make1(HnswEntryCandidate(entryPoint, q, index, procinfo, collation, false));
for (int lc = entryPoint->level; lc >= 1; lc--)
{
w = HnswSearchLayer(base, q, ep, 1, lc, index, procinfo, collation, m, false, NULL);
w = HnswSearchLayer(q, ep, 1, lc, index, procinfo, collation, m, false, NULL);
ep = w;
}
return HnswSearchLayer(base, q, ep, hnsw_ef_search, 0, index, procinfo, collation, m, false, NULL);
return HnswSearchLayer(q, ep, hnsw_ef_search, 0, index, procinfo, collation, m, false, NULL);
}
/*
@@ -185,24 +184,26 @@ hnswgettuple(IndexScanDesc scan, ScanDirection dir)
while (list_length(so->w) > 0)
{
char *base = NULL;
HnswCandidate *hc = llast(so->w);
HnswElement element = HnswPtrAccess(base, hc->element);
ItemPointer heaptid;
/* Move to next element if no valid heap TIDs */
if (element->heaptidsLength == 0)
if (hc->element->heaptidsLength == 0)
{
so->w = list_delete_last(so->w);
continue;
}
heaptid = &element->heaptids[--element->heaptidsLength];
heaptid = &hc->element->heaptids[--hc->element->heaptidsLength];
MemoryContextSwitchTo(oldCtx);
#if PG_VERSION_NUM >= 120000
scan->xs_heaptid = *heaptid;
scan->xs_recheck = false;
#else
scan->xs_ctup.t_self = *heaptid;
#endif
scan->xs_recheckorderby = false;
return true;
}

View File

@@ -2,13 +2,9 @@
#include <math.h>
#include "access/generic_xlog.h"
#include "hnsw.h"
#include "lib/pairingheap.h"
#include "storage/bufmgr.h"
#include "utils/datum.h"
#include "utils/memdebug.h"
#include "utils/rel.h"
#include "vector.h"
#if PG_VERSION_NUM >= 130000
@@ -60,6 +56,13 @@ hash_tid(ItemPointerData tid)
#define SH_DEFINE
#include "lib/simplehash.h"
/* Needed to include simplehash.h twice */
#if PG_VERSION_NUM < 120000
#undef SH_EQUAL
#define sh_log2 pointerhash_sh_log2
#define sh_pow2 pointerhash_sh_pow2
#endif
/* Pointer hash table */
static uint32
hash_pointer(uintptr_t ptr)
@@ -81,31 +84,9 @@ hash_pointer(uintptr_t ptr)
#define SH_DEFINE
#include "lib/simplehash.h"
/* Offset hash table */
static uint32
hash_offset(Size offset)
{
#if SIZEOF_SIZE_T == 8
return murmurhash64((uint64) offset);
#else
return murmurhash32((uint32) offset);
#endif
}
#define SH_PREFIX offsethash
#define SH_ELEMENT_TYPE OffsetHashEntry
#define SH_KEY_TYPE Size
#define SH_KEY offset
#define SH_HASH_KEY(tb, key) hash_offset(key)
#define SH_EQUAL(tb, a, b) (a == b)
#define SH_SCOPE extern
#define SH_DEFINE
#include "lib/simplehash.h"
typedef union
{
pointerhash_hash *pointers;
offsethash_hash *offsets;
tidhash_hash *tids;
} visited_hash;
@@ -203,53 +184,35 @@ HnswInitPage(Buffer buf, Page page)
HnswPageGetOpaque(page)->page_id = HNSW_PAGE_ID;
}
/*
* Allocate a neighbor array
*/
static HnswNeighborArray *
HnswInitNeighborArray(int lm, HnswAllocator * allocator)
{
HnswNeighborArray *a = HnswAlloc(allocator, HNSW_NEIGHBOR_ARRAY_SIZE(lm));
a->length = 0;
a->closerSet = false;
return a;
}
/*
* Allocate neighbors
*/
void
HnswInitNeighbors(char *base, HnswElement element, int m, HnswAllocator * allocator)
HnswInitNeighbors(HnswElement element, int m)
{
int level = element->level;
HnswNeighborArrayPtr *neighborList = (HnswNeighborArrayPtr *) HnswAlloc(allocator, sizeof(HnswNeighborArrayPtr) * (level + 1));
HnswPtrStore(base, element->neighbors, neighborList);
element->neighbors = palloc(sizeof(HnswNeighborArray) * (level + 1));
for (int lc = 0; lc <= level; lc++)
HnswPtrStore(base, neighborList[lc], HnswInitNeighborArray(HnswGetLayerM(m, lc), allocator));
}
/*
* Allocate memory from the allocator
*/
void *
HnswAlloc(HnswAllocator * allocator, Size size)
{
if (allocator)
return (*(allocator)->alloc) (size, (allocator)->state);
HnswNeighborArray *a;
int lm = HnswGetLayerM(m, lc);
return palloc(size);
a = &element->neighbors[lc];
a->length = 0;
a->items = palloc(sizeof(HnswCandidate) * lm);
a->closerSet = false;
}
}
/*
* Allocate an element
*/
HnswElement
HnswInitElement(char *base, ItemPointer heaptid, int m, double ml, int maxLevel, HnswAllocator * allocator)
HnswInitElement(ItemPointer heaptid, int m, double ml, int maxLevel)
{
HnswElement element = HnswAlloc(allocator, sizeof(HnswElementData));
HnswElement element = palloc(sizeof(HnswElementData));
int level = (int) (-log(RandomDouble()) * ml);
@@ -263,9 +226,9 @@ HnswInitElement(char *base, ItemPointer heaptid, int m, double ml, int maxLevel,
element->level = level;
element->deleted = 0;
HnswInitNeighbors(base, element, m, allocator);
HnswInitNeighbors(element, m);
HnswPtrStore(base, element->value, (Pointer) NULL);
element->value = PointerGetDatum(NULL);
return element;
}
@@ -286,12 +249,11 @@ HnswElement
HnswInitElementFromBlock(BlockNumber blkno, OffsetNumber offno)
{
HnswElement element = palloc(sizeof(HnswElementData));
char *base = NULL;
element->blkno = blkno;
element->offno = offno;
HnswPtrStore(base, element->neighbors, (HnswNeighborArrayPtr *) NULL);
HnswPtrStore(base, element->value, (Pointer) NULL);
element->neighbors = NULL;
element->value = PointerGetDatum(NULL);
return element;
}
@@ -316,10 +278,7 @@ HnswGetMetaPageInfo(Relation index, int *m, HnswElement * entryPoint)
if (entryPoint != NULL)
{
if (BlockNumberIsValid(metap->entryBlkno))
{
*entryPoint = HnswInitElementFromBlock(metap->entryBlkno, metap->entryOffno);
(*entryPoint)->level = metap->entryLevel;
}
else
*entryPoint = NULL;
}
@@ -404,10 +363,8 @@ HnswUpdateMetaPage(Relation index, int updateEntry, HnswElement entryPoint, Bloc
* Set element tuple, except for neighbor info
*/
void
HnswSetElementTuple(char *base, HnswElementTuple etup, HnswElement element)
HnswSetElementTuple(HnswElementTuple etup, HnswElement element)
{
Pointer valuePtr = HnswPtrAccess(base, element->value);
etup->type = HNSW_ELEMENT_TUPLE_TYPE;
etup->level = element->level;
etup->deleted = 0;
@@ -418,14 +375,14 @@ HnswSetElementTuple(char *base, HnswElementTuple etup, HnswElement element)
else
ItemPointerSetInvalid(&etup->heaptids[i]);
}
memcpy(&etup->data, valuePtr, VARSIZE_ANY(valuePtr));
memcpy(&etup->data, DatumGetPointer(element->value), VARSIZE_ANY(DatumGetPointer(element->value)));
}
/*
* Set neighbor tuple
*/
void
HnswSetNeighborTuple(char *base, HnswNeighborTuple ntup, HnswElement e, int m)
HnswSetNeighborTuple(HnswNeighborTuple ntup, HnswElement e, int m)
{
int idx = 0;
@@ -433,7 +390,7 @@ HnswSetNeighborTuple(char *base, HnswNeighborTuple ntup, HnswElement e, int m)
for (int lc = e->level; lc >= 0; lc--)
{
HnswNeighborArray *neighbors = HnswGetNeighbors(base, e, lc);
HnswNeighborArray *neighbors = HnswGetNeighbors(e, lc);
int lm = HnswGetLayerM(m, lc);
for (int i = 0; i < lm; i++)
@@ -443,9 +400,8 @@ HnswSetNeighborTuple(char *base, HnswNeighborTuple ntup, HnswElement e, int m)
if (i < neighbors->length)
{
HnswCandidate *hc = &neighbors->items[i];
HnswElement hce = HnswPtrAccess(base, hc->element);
ItemPointerSet(indextid, hce->blkno, hce->offno);
ItemPointerSet(indextid, hc->element->blkno, hc->element->offno);
}
else
ItemPointerSetInvalid(indextid);
@@ -461,14 +417,12 @@ HnswSetNeighborTuple(char *base, HnswNeighborTuple ntup, HnswElement e, int m)
static void
LoadNeighborsFromPage(HnswElement element, Relation index, Page page, int m)
{
char *base = NULL;
HnswNeighborTuple ntup = (HnswNeighborTuple) PageGetItem(page, PageGetItemId(page, element->neighborOffno));
int neighborCount = (element->level + 2) * m;
Assert(HnswIsNeighborTuple(ntup));
HnswInitNeighbors(base, element, m, NULL);
HnswInitNeighbors(element, m);
/* Ensure expected neighbors */
if (ntup->count != neighborCount)
@@ -494,9 +448,9 @@ LoadNeighborsFromPage(HnswElement element, Relation index, Page page, int m)
if (level < 0)
level = 0;
neighbors = HnswGetNeighbors(base, element, level);
neighbors = HnswGetNeighbors(element, level);
hc = &neighbors->items[neighbors->length++];
HnswPtrStore(base, hc->element, e);
hc->element = e;
}
}
@@ -543,12 +497,7 @@ HnswLoadElementFromTuple(HnswElement element, HnswElementTuple etup, bool loadHe
}
if (loadVec)
{
char *base = NULL;
Datum value = datumCopy(PointerGetDatum(&etup->data), false, -1);
HnswPtrStore(base, element->value, DatumGetPointer(value));
}
element->value = datumCopy(PointerGetDatum(&etup->data), false, -1);
}
/*
@@ -584,27 +533,24 @@ HnswLoadElement(HnswElement element, float *distance, Datum *q, Relation index,
* Get the distance for a candidate
*/
static float
GetCandidateDistance(char *base, HnswCandidate * hc, Datum q, FmgrInfo *procinfo, Oid collation)
GetCandidateDistance(HnswCandidate * hc, Datum q, FmgrInfo *procinfo, Oid collation)
{
HnswElement hce = HnswPtrAccess(base, hc->element);
Datum value = HnswGetValue(base, hce);
return DatumGetFloat8(FunctionCall2Coll(procinfo, collation, q, value));
return DatumGetFloat8(FunctionCall2Coll(procinfo, collation, q, hc->element->value));
}
/*
* Create a candidate for the entry point
*/
HnswCandidate *
HnswEntryCandidate(char *base, HnswElement entryPoint, Datum q, Relation index, FmgrInfo *procinfo, Oid collation, bool loadVec)
HnswEntryCandidate(HnswElement entryPoint, Datum q, Relation index, FmgrInfo *procinfo, Oid collation, bool loadVec)
{
HnswCandidate *hc = palloc(sizeof(HnswCandidate));
HnswPtrStore(base, hc->element, entryPoint);
hc->element = entryPoint;
if (index == NULL)
hc->distance = GetCandidateDistance(base, hc, q, procinfo, collation);
hc->distance = GetCandidateDistance(hc, q, procinfo, collation);
else
HnswLoadElement(entryPoint, &hc->distance, &q, index, procinfo, collation, loadVec);
HnswLoadElement(hc->element, &hc->distance, &q, index, procinfo, collation, loadVec);
return hc;
}
@@ -650,79 +596,34 @@ CreatePairingHeapNode(HnswCandidate * c)
return node;
}
/*
* Init visited
*/
static inline void
InitVisited(char *base, visited_hash * v, Relation index, int ef, int m)
{
if (index != NULL)
v->tids = tidhash_create(CurrentMemoryContext, ef * m * 2, NULL);
else if (base != NULL)
v->offsets = offsethash_create(CurrentMemoryContext, ef * m * 2, NULL);
else
v->pointers = pointerhash_create(CurrentMemoryContext, ef * m * 2, NULL);
}
/*
* Add to visited
*/
static inline void
AddToVisited(char *base, visited_hash * v, HnswCandidate * hc, Relation index, bool *found)
AddToVisited(visited_hash v, HnswCandidate * hc, Relation index, bool *found)
{
if (index != NULL)
{
HnswElement element = HnswPtrAccess(base, hc->element);
ItemPointerData indextid;
ItemPointerSet(&indextid, element->blkno, element->offno);
tidhash_insert(v->tids, indextid, found);
}
else if (base != NULL)
if (index == NULL)
{
#if PG_VERSION_NUM >= 130000
HnswElement element = HnswPtrAccess(base, hc->element);
offsethash_insert_hash(v->offsets, HnswPtrOffset(hc->element), element->hash, found);
pointerhash_insert_hash(v.pointers, (uintptr_t) hc->element, hc->element->hash, found);
#else
offsethash_insert(v->offsets, HnswPtrOffset(hc->element), found);
pointerhash_insert(v.pointers, (uintptr_t) hc->element, found);
#endif
}
else
{
#if PG_VERSION_NUM >= 130000
HnswElement element = HnswPtrAccess(base, hc->element);
ItemPointerData indextid;
pointerhash_insert_hash(v->pointers, (uintptr_t) HnswPtrPointer(hc->element), element->hash, found);
#else
pointerhash_insert(v->pointers, (uintptr_t) HnswPtrPointer(hc->element), found);
#endif
ItemPointerSet(&indextid, hc->element->blkno, hc->element->offno);
tidhash_insert(v.tids, indextid, found);
}
}
/*
* Count element towards ef
*/
static inline bool
CountElement(char *base, HnswElement skipElement, HnswCandidate * hc)
{
HnswElement e;
if (skipElement == NULL)
return true;
/* Ensure does not access heaptidsLength during in-memory build */
pg_memory_barrier();
e = HnswPtrAccess(base, hc->element);
return e->heaptidsLength != 0;
}
/*
* Algorithm 2 from paper
*/
List *
HnswSearchLayer(char *base, 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, int m, bool inserting, HnswElement skipElement)
{
List *w = NIL;
pairingheap *C = pairingheap_allocate(CompareNearestCandidates, NULL);
@@ -730,17 +631,12 @@ HnswSearchLayer(char *base, Datum q, List *ep, int ef, int lc, Relation index, F
int wlen = 0;
visited_hash v;
ListCell *lc2;
HnswNeighborArray *neighborhoodData = NULL;
Size neighborhoodSize;
InitVisited(base, &v, index, ef, m);
/* Create local memory for neighborhood if needed */
/* Create hash table */
if (index == NULL)
{
neighborhoodSize = HNSW_NEIGHBOR_ARRAY_SIZE(HnswGetLayerM(m, lc));
neighborhoodData = palloc(neighborhoodSize);
}
v.pointers = pointerhash_create(CurrentMemoryContext, ef * m * 2, NULL);
else
v.tids = tidhash_create(CurrentMemoryContext, ef * m * 2, NULL);
/* Add entry points to v, C, and W */
foreach(lc2, ep)
@@ -748,7 +644,7 @@ HnswSearchLayer(char *base, Datum q, List *ep, int ef, int lc, Relation index, F
HnswCandidate *hc = (HnswCandidate *) lfirst(lc2);
bool found;
AddToVisited(base, &v, hc, index, &found);
AddToVisited(v, hc, index, &found);
pairingheap_add(C, &(CreatePairingHeapNode(hc)->ph_node));
pairingheap_add(W, &(CreatePairingHeapNode(hc)->ph_node));
@@ -758,7 +654,7 @@ HnswSearchLayer(char *base, Datum q, List *ep, int ef, int lc, Relation index, F
* would be ideal to do this for inserts as well, but this could
* affect insert performance.
*/
if (CountElement(base, skipElement, hc))
if (skipElement == NULL || hc->element->heaptidsLength != 0)
wlen++;
}
@@ -767,51 +663,38 @@ HnswSearchLayer(char *base, Datum q, List *ep, int ef, int lc, Relation index, F
HnswNeighborArray *neighborhood;
HnswCandidate *c = ((HnswPairingHeapNode *) pairingheap_remove_first(C))->inner;
HnswCandidate *f = ((HnswPairingHeapNode *) pairingheap_first(W))->inner;
HnswElement cElement;
if (c->distance > f->distance)
break;
cElement = HnswPtrAccess(base, c->element);
if (HnswPtrIsNull(base, cElement->neighbors))
HnswLoadNeighbors(cElement, index, m);
if (c->element->neighbors == NULL)
HnswLoadNeighbors(c->element, index, m);
/* Get the neighborhood at layer lc */
neighborhood = HnswGetNeighbors(base, cElement, lc);
/* Copy neighborhood to local memory if needed */
if (index == NULL)
{
LWLockAcquire(&cElement->lock, LW_SHARED);
memcpy(neighborhoodData, neighborhood, neighborhoodSize);
LWLockRelease(&cElement->lock);
neighborhood = neighborhoodData;
}
neighborhood = HnswGetNeighbors(c->element, lc);
for (int i = 0; i < neighborhood->length; i++)
{
HnswCandidate *e = &neighborhood->items[i];
bool visited;
AddToVisited(base, &v, e, index, &visited);
AddToVisited(v, e, index, &visited);
if (!visited)
{
float eDistance;
HnswElement eElement = HnswPtrAccess(base, e->element);
f = ((HnswPairingHeapNode *) pairingheap_first(W))->inner;
if (index == NULL)
eDistance = GetCandidateDistance(base, e, q, procinfo, collation);
eDistance = GetCandidateDistance(e, q, procinfo, collation);
else
HnswLoadElement(eElement, &eDistance, &q, index, procinfo, collation, inserting);
HnswLoadElement(e->element, &eDistance, &q, index, procinfo, collation, inserting);
Assert(!eElement->deleted);
Assert(!e->element->deleted);
/* Make robust to issues */
if (eElement->level < lc)
if (e->element->level < lc)
continue;
if (eDistance < f->distance || wlen < ef)
@@ -819,7 +702,7 @@ HnswSearchLayer(char *base, Datum q, List *ep, int ef, int lc, Relation index, F
/* Copy e */
HnswCandidate *ec = palloc(sizeof(HnswCandidate));
HnswPtrStore(base, ec->element, eElement);
ec->element = e->element;
ec->distance = eDistance;
pairingheap_add(C, &(CreatePairingHeapNode(ec)->ph_node));
@@ -830,7 +713,7 @@ HnswSearchLayer(char *base, Datum q, List *ep, int ef, int lc, Relation index, F
* vacuuming. It would be ideal to do this for inserts as
* well, but this could affect insert performance.
*/
if (CountElement(base, skipElement, e))
if (skipElement == NULL || e->element->heaptidsLength != 0)
{
wlen++;
@@ -855,7 +738,7 @@ HnswSearchLayer(char *base, Datum q, List *ep, int ef, int lc, Relation index, F
}
/*
* Compare candidate distances with pointer tie-breaker
* Compare candidate distances
*/
static int
#if PG_VERSION_NUM >= 130000
@@ -873,38 +756,10 @@ CompareCandidateDistances(const void *a, const void *b)
if (hca->distance > hcb->distance)
return -1;
if (HnswPtrPointer(hca->element) < HnswPtrPointer(hcb->element))
if (hca->element < hcb->element)
return 1;
if (HnswPtrPointer(hca->element) > HnswPtrPointer(hcb->element))
return -1;
return 0;
}
/*
* Compare candidate distances with offset tie-breaker
*/
static int
#if PG_VERSION_NUM >= 130000
CompareCandidateDistancesOffset(const ListCell *a, const ListCell *b)
#else
CompareCandidateDistancesOffset(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 (HnswPtrOffset(hca->element) < HnswPtrOffset(hcb->element))
return 1;
if (HnswPtrOffset(hca->element) > HnswPtrOffset(hcb->element))
if (hca->element > hcb->element)
return -1;
return 0;
@@ -914,28 +769,46 @@ CompareCandidateDistancesOffset(const void *a, const void *b)
* Calculate the distance between elements
*/
static float
HnswGetDistance(char *base, HnswElement a, HnswElement b, FmgrInfo *procinfo, Oid collation)
HnswGetDistance(HnswElement a, HnswElement b, int lc, FmgrInfo *procinfo, Oid collation)
{
Datum aValue = HnswGetValue(base, a);
Datum bValue = HnswGetValue(base, b);
/* Look for cached distance */
if (a->neighbors != NULL)
{
HnswNeighborArray *neighbors = HnswGetNeighbors(a, lc);
return DatumGetFloat8(FunctionCall2Coll(procinfo, collation, aValue, bValue));
for (int i = 0; i < neighbors->length; i++)
{
if (neighbors->items[i].element == b)
return neighbors->items[i].distance;
}
}
if (b->neighbors != NULL)
{
HnswNeighborArray *neighbors = HnswGetNeighbors(b, lc);
for (int i = 0; i < neighbors->length; i++)
{
if (neighbors->items[i].element == a)
return neighbors->items[i].distance;
}
}
return DatumGetFloat8(FunctionCall2Coll(procinfo, collation, a->value, b->value));
}
/*
* Check if an element is closer to q than any element from R
*/
static bool
CheckElementCloser(char *base, HnswCandidate * e, List *r, FmgrInfo *procinfo, Oid collation)
CheckElementCloser(HnswCandidate * e, List *r, int lc, FmgrInfo *procinfo, Oid collation)
{
HnswElement eElement = HnswPtrAccess(base, e->element);
ListCell *lc2;
foreach(lc2, r)
{
HnswCandidate *ri = lfirst(lc2);
HnswElement riElement = HnswPtrAccess(base, ri->element);
float distance = HnswGetDistance(base, eElement, riElement, procinfo, collation);
float distance = HnswGetDistance(e->element, ri->element, lc, procinfo, collation);
if (distance <= e->distance)
return false;
@@ -948,31 +821,26 @@ CheckElementCloser(char *base, HnswCandidate * e, List *r, FmgrInfo *procinfo, O
* Algorithm 4 from paper
*/
static List *
SelectNeighbors(char *base, List *c, int lm, 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, HnswElement e2, HnswCandidate * newCandidate, HnswCandidate * *pruned, bool sortCandidates)
{
List *r = NIL;
List *w = list_copy(c);
pairingheap *wd;
HnswNeighborArray *neighbors = HnswGetNeighbors(base, e2, lc);
HnswNeighborArray *neighbors = HnswGetNeighbors(e2, lc);
bool mustCalculate = !neighbors->closerSet;
List *added = NIL;
bool removedAny = false;
if (list_length(w) <= lm)
if (list_length(w) <= m)
return w;
wd = pairingheap_allocate(CompareNearestCandidates, NULL);
/* Ensure order of candidates is deterministic for closer caching */
if (sortCandidates)
{
if (base == NULL)
list_sort(w, CompareCandidateDistances);
else
list_sort(w, CompareCandidateDistancesOffset);
}
while (list_length(w) > 0 && list_length(r) < lm)
while (list_length(w) > 0 && list_length(r) < m)
{
/* Assumes w is already ordered desc */
HnswCandidate *e = llast(w);
@@ -981,20 +849,16 @@ SelectNeighbors(char *base, List *c, int lm, int lc, FmgrInfo *procinfo, Oid col
/* Use previous state of r and wd to skip work when possible */
if (mustCalculate)
e->closer = CheckElementCloser(base, e, r, procinfo, collation);
e->closer = CheckElementCloser(e, r, lc, procinfo, collation);
else if (list_length(added) > 0)
{
/* Keep Valgrind happy for in-memory, parallel builds */
if (base != NULL)
VALGRIND_MAKE_MEM_DEFINED(&e->closer, 1);
/*
* 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(base, e, added, procinfo, collation);
e->closer = CheckElementCloser(e, added, lc, procinfo, collation);
if (!e->closer)
removedAny = true;
@@ -1007,7 +871,7 @@ SelectNeighbors(char *base, List *c, int lm, int lc, FmgrInfo *procinfo, Oid col
*/
if (removedAny)
{
e->closer = CheckElementCloser(base, e, r, procinfo, collation);
e->closer = CheckElementCloser(e, r, lc, procinfo, collation);
if (e->closer)
added = lappend(added, e);
}
@@ -1015,15 +879,11 @@ SelectNeighbors(char *base, List *c, int lm, int lc, FmgrInfo *procinfo, Oid col
}
else if (e == newCandidate)
{
e->closer = CheckElementCloser(base, e, r, procinfo, collation);
e->closer = CheckElementCloser(e, r, lc, procinfo, collation);
if (e->closer)
added = lappend(added, e);
}
/* Keep Valgrind happy for in-memory, parallel builds */
if (base != NULL)
VALGRIND_MAKE_MEM_DEFINED(&e->closer, 1);
if (e->closer)
r = lappend(r, e);
else
@@ -1034,7 +894,7 @@ SelectNeighbors(char *base, List *c, int lm, int lc, FmgrInfo *procinfo, Oid col
neighbors->closerSet = sortCandidates;
/* Keep pruned connections */
while (!pairingheap_is_empty(wd) && list_length(r) < lm)
while (!pairingheap_is_empty(wd) && list_length(r) < m)
r = lappend(r, ((HnswPairingHeapNode *) pairingheap_remove_first(wd))->inner);
/* Return pruned for update connections */
@@ -1053,10 +913,10 @@ SelectNeighbors(char *base, List *c, int lm, int lc, FmgrInfo *procinfo, Oid col
* Add connections
*/
static void
AddConnections(char *base, HnswElement element, List *neighbors, int lc)
AddConnections(HnswElement element, List *neighbors, int m, int lc)
{
ListCell *lc2;
HnswNeighborArray *a = HnswGetNeighbors(base, element, lc);
HnswNeighborArray *a = HnswGetNeighbors(element, lc);
foreach(lc2, neighbors)
a->items[a->length++] = *((HnswCandidate *) lfirst(lc2));
@@ -1066,16 +926,16 @@ AddConnections(char *base, HnswElement element, List *neighbors, int lc)
* Update connections
*/
void
HnswUpdateConnection(char *base, HnswElement element, HnswCandidate * hc, int lm, int lc, int *updateIdx, Relation index, FmgrInfo *procinfo, Oid collation)
HnswUpdateConnection(HnswElement element, HnswCandidate * hc, int m, int lc, int *updateIdx, Relation index, FmgrInfo *procinfo, Oid collation)
{
HnswElement hce = HnswPtrAccess(base, hc->element);
HnswNeighborArray *currentNeighbors = HnswGetNeighbors(base, hce, lc);
HnswNeighborArray *currentNeighbors = HnswGetNeighbors(hc->element, lc);
HnswCandidate hc2;
HnswPtrStore(base, hc2.element, element);
hc2.element = element;
hc2.distance = hc->distance;
if (currentNeighbors->length < lm)
if (currentNeighbors->length < m)
{
currentNeighbors->items[currentNeighbors->length++] = hc2;
@@ -1091,20 +951,19 @@ HnswUpdateConnection(char *base, HnswElement element, HnswCandidate * hc, int lm
/* Load elements on insert */
if (index != NULL)
{
Datum q = HnswGetValue(base, hce);
Datum q = hc->element->value;
for (int i = 0; i < currentNeighbors->length; i++)
{
HnswCandidate *hc3 = &currentNeighbors->items[i];
HnswElement hc3Element = HnswPtrAccess(base, hc3->element);
if (HnswPtrIsNull(base, hc3Element->value))
HnswLoadElement(hc3Element, &hc3->distance, &q, index, procinfo, collation, true);
if (DatumGetPointer(hc3->element->value) == NULL)
HnswLoadElement(hc3->element, &hc3->distance, &q, index, procinfo, collation, true);
else
hc3->distance = GetCandidateDistance(base, hc3, q, procinfo, collation);
hc3->distance = GetCandidateDistance(hc3, q, procinfo, collation);
/* Prune element if being deleted */
if (hc3Element->heaptidsLength == 0)
if (hc3->element->heaptidsLength == 0)
{
pruned = &currentNeighbors->items[i];
break;
@@ -1121,7 +980,7 @@ HnswUpdateConnection(char *base, HnswElement element, HnswCandidate * hc, int lm
c = lappend(c, &currentNeighbors->items[i]);
c = lappend(c, &hc2);
SelectNeighbors(base, c, lm, lc, procinfo, collation, hce, &hc2, &pruned, true);
SelectNeighbors(c, m, lc, procinfo, collation, hc->element, &hc2, &pruned, true);
/* Should not happen */
if (pruned == NULL)
@@ -1131,7 +990,7 @@ HnswUpdateConnection(char *base, HnswElement element, HnswCandidate * hc, int lm
/* Find and replace the pruned element */
for (int i = 0; i < currentNeighbors->length; i++)
{
if (HnswPtrEqual(base, currentNeighbors->items[i].element, pruned->element))
if (currentNeighbors->items[i].element == pruned->element)
{
currentNeighbors->items[i] = hc2;
@@ -1149,65 +1008,43 @@ HnswUpdateConnection(char *base, HnswElement element, HnswCandidate * hc, int lm
* Remove elements being deleted or skipped
*/
static List *
RemoveElements(char *base, List *w, HnswElement skipElement)
RemoveElements(List *w, HnswElement skipElement)
{
ListCell *lc2;
List *w2 = NIL;
/* Ensure does not access heaptidsLength during in-memory build */
pg_memory_barrier();
foreach(lc2, w)
{
HnswCandidate *hc = (HnswCandidate *) lfirst(lc2);
HnswElement hce = HnswPtrAccess(base, hc->element);
/* Skip self for vacuuming update */
if (skipElement != NULL && hce->blkno == skipElement->blkno && hce->offno == skipElement->offno)
if (skipElement != NULL && hc->element->blkno == skipElement->blkno && hc->element->offno == skipElement->offno)
continue;
if (hce->heaptidsLength != 0)
if (hc->element->heaptidsLength != 0)
w2 = lappend(w2, hc);
}
return w2;
}
#if PG_VERSION_NUM >= 130000
/*
* Precompute hash
*/
static void
PrecomputeHash(char *base, HnswElement element)
{
HnswElementPtr ptr;
HnswPtrStore(base, ptr, element);
if (base == NULL)
element->hash = hash_pointer((uintptr_t) HnswPtrPointer(ptr));
else
element->hash = hash_offset(HnswPtrOffset(ptr));
}
#endif
/*
* Algorithm 1 from paper
*/
void
HnswFindElementNeighbors(char *base, HnswElement element, HnswElement entryPoint, Relation index, FmgrInfo *procinfo, Oid collation, int m, int efConstruction, bool existing)
HnswInsertElement(HnswElement element, HnswElement entryPoint, Relation index, FmgrInfo *procinfo, Oid collation, int m, int efConstruction, bool existing)
{
List *ep;
List *w;
int level = element->level;
int entryLevel;
Datum q = HnswGetValue(base, element);
Datum q = element->value;
HnswElement skipElement = existing ? element : NULL;
#if PG_VERSION_NUM >= 130000
/* Precompute hash */
if (index == NULL)
PrecomputeHash(base, element);
element->hash = hash_pointer((uintptr_t) element);
#endif
/* No neighbors if no entry point */
@@ -1215,13 +1052,13 @@ HnswFindElementNeighbors(char *base, HnswElement element, HnswElement entryPoint
return;
/* Get entry point and level */
ep = list_make1(HnswEntryCandidate(base, entryPoint, q, index, procinfo, collation, true));
ep = list_make1(HnswEntryCandidate(entryPoint, q, index, procinfo, collation, true));
entryLevel = entryPoint->level;
/* 1st phase: greedy search to insert level */
for (int lc = entryLevel; lc >= level + 1; lc--)
{
w = HnswSearchLayer(base, q, ep, 1, lc, index, procinfo, collation, m, true, skipElement);
w = HnswSearchLayer(q, ep, 1, lc, index, procinfo, collation, m, true, skipElement);
ep = w;
}
@@ -1239,12 +1076,12 @@ HnswFindElementNeighbors(char *base, HnswElement element, HnswElement entryPoint
List *neighbors;
List *lw;
w = HnswSearchLayer(base, q, ep, efConstruction, lc, index, procinfo, collation, m, true, skipElement);
w = HnswSearchLayer(q, ep, efConstruction, lc, index, procinfo, collation, m, true, skipElement);
/* Elements being deleted or skipped can help with search */
/* but should be removed before selecting neighbors */
if (index != NULL)
lw = RemoveElements(base, w, skipElement);
lw = RemoveElements(w, skipElement);
else
lw = w;
@@ -1253,9 +1090,9 @@ HnswFindElementNeighbors(char *base, HnswElement element, HnswElement entryPoint
* sortCandidates to true for in-memory builds to enable closer
* caching, but there does not seem to be a difference in performance.
*/
neighbors = SelectNeighbors(base, lw, lm, lc, procinfo, collation, element, NULL, NULL, false);
neighbors = SelectNeighbors(lw, lm, lc, procinfo, collation, element, NULL, NULL, false);
AddConnections(base, element, neighbors, lc);
AddConnections(element, neighbors, lm, lc);
ep = w;
}

View File

@@ -2,7 +2,6 @@
#include <math.h>
#include "access/generic_xlog.h"
#include "commands/vacuum.h"
#include "hnsw.h"
#include "storage/bufmgr.h"
@@ -60,7 +59,8 @@ RemoveHeapTids(HnswVacuumState * vacuumstate)
/* Iterate over nodes */
for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
{
HnswElementTuple etup = (HnswElementTuple) PageGetItem(page, PageGetItemId(page, offno));
ItemId itemid = PageGetItemId(page, offno);
HnswElementTuple etup = (HnswElementTuple) PageGetItem(page, itemid);
int idx = 0;
bool itemUpdated = false;
@@ -91,10 +91,15 @@ RemoveHeapTids(HnswVacuumState * vacuumstate)
if (itemUpdated)
{
Size etupSize = ItemIdGetLength(itemid);
/* Mark rest as invalid */
for (int i = idx; i < HNSW_HEAPTIDS; i++)
ItemPointerSetInvalid(&etup->heaptids[i]);
if (!PageIndexTupleOverwrite(page, offno, (Item) etup, etupSize))
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
updated = true;
}
}
@@ -194,25 +199,21 @@ RepairGraphElement(HnswVacuumState * vacuumstate, HnswElement element, HnswEleme
BufferAccessStrategy bas = vacuumstate->bas;
HnswNeighborTuple ntup = vacuumstate->ntup;
Size ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(element->level, m);
char *base = NULL;
/* Skip if element is entry point */
if (entryPoint != NULL && element->blkno == entryPoint->blkno && element->offno == entryPoint->offno)
return;
/* Init fields */
HnswInitNeighbors(base, element, m, NULL);
HnswInitNeighbors(element, m);
element->heaptidsLength = 0;
/* Find neighbors for element, skipping itself */
HnswFindElementNeighbors(base, element, entryPoint, index, procinfo, collation, m, efConstruction, true);
/* Zero memory for each element */
MemSet(ntup, 0, HNSW_TUPLE_ALLOC_SIZE);
/* Add element to graph, skipping itself */
HnswInsertElement(element, entryPoint, index, procinfo, collation, m, efConstruction, true);
/* Update neighbor tuple */
/* Do this before getting page to minimize locking */
HnswSetNeighborTuple(base, ntup, element, m);
HnswSetNeighborTuple(ntup, element, m);
/* Get neighbor page */
buf = ReadBufferExtended(index, MAIN_FORKNUM, element->neighborPage, RBM_NORMAL, bas);
@@ -229,7 +230,7 @@ RepairGraphElement(HnswVacuumState * vacuumstate, HnswElement element, HnswEleme
UnlockReleaseBuffer(buf);
/* Update neighbors */
HnswUpdateNeighborsOnDisk(index, procinfo, collation, element, m, true, false);
HnswUpdateNeighborPages(index, procinfo, collation, element, m, true, false);
}
/*
@@ -300,7 +301,7 @@ RepairGraphEntryPoint(HnswVacuumState * vacuumstate)
{
/* Reset neighbors from previous update */
if (highestPoint != NULL)
HnswPtrStore((char *) NULL, highestPoint->neighbors, (HnswNeighborArrayPtr *) NULL);
highestPoint->neighbors = NULL;
RepairGraphElement(vacuumstate, entryPoint, highestPoint);
}
@@ -476,8 +477,11 @@ MarkDeleted(HnswVacuumState * vacuumstate)
/* Update element and neighbors together */
for (offno = FirstOffsetNumber; offno <= maxoffno; offno = OffsetNumberNext(offno))
{
HnswElementTuple etup = (HnswElementTuple) PageGetItem(page, PageGetItemId(page, offno));
ItemId itemid = PageGetItemId(page, offno);
HnswElementTuple etup = (HnswElementTuple) PageGetItem(page, itemid);
HnswNeighborTuple ntup;
Size etupSize;
Size ntupSize;
Buffer nbuf;
Page npage;
BlockNumber neighborPage;
@@ -501,6 +505,10 @@ MarkDeleted(HnswVacuumState * vacuumstate)
if (ItemPointerIsValid(&etup->heaptids[0]))
continue;
/* Calculate sizes */
etupSize = ItemIdGetLength(itemid);
ntupSize = HNSW_NEIGHBOR_TUPLE_SIZE(etup->level, vacuumstate->m);
/* Get neighbor page */
neighborPage = ItemPointerGetBlockNumber(&etup->neighbortid);
neighborOffno = ItemPointerGetOffsetNumber(&etup->neighbortid);
@@ -527,10 +535,13 @@ MarkDeleted(HnswVacuumState * vacuumstate)
for (int i = 0; i < ntup->count; i++)
ItemPointerSetInvalid(&ntup->indextids[i]);
/*
* We modified the tuples in place, no need to call
* PageIndexTupleOverwrite
*/
/* Overwrite element tuple */
if (!PageIndexTupleOverwrite(page, offno, (Item) etup, etupSize))
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
/* Overwrite neighbor tuple */
if (!PageIndexTupleOverwrite(npage, neighborOffno, (Item) ntup, ntupSize))
elog(ERROR, "failed to add index item to \"%s\"", RelationGetRelationName(index));
/* Commit */
GenericXLogFinish(state);
@@ -575,7 +586,7 @@ InitVacuumState(HnswVacuumState * vacuumstate, IndexVacuumInfo *info, IndexBulkD
vacuumstate->bas = GetAccessStrategy(BAS_BULKREAD);
vacuumstate->procinfo = index_getprocinfo(index, 1, HNSW_DISTANCE_PROC);
vacuumstate->collation = index->rd_indcollation[0];
vacuumstate->ntup = palloc0(HNSW_TUPLE_ALLOC_SIZE);
vacuumstate->ntup = palloc0(BLCKSZ);
vacuumstate->tmpCtx = AllocSetContextCreate(CurrentMemoryContext,
"Hnsw vacuum temporary context",
ALLOCSET_DEFAULT_SIZES);

View File

@@ -2,38 +2,58 @@
#include <float.h>
#include "access/table.h"
#include "access/tableam.h"
#include "access/parallel.h"
#include "access/xact.h"
#include "catalog/index.h"
#include "catalog/pg_operator_d.h"
#include "catalog/pg_type_d.h"
#include "commands/progress.h"
#include "ivfflat.h"
#include "miscadmin.h"
#include "optimizer/optimizer.h"
#include "storage/bufmgr.h"
#include "tcop/tcopprot.h"
#include "utils/memutils.h"
#if PG_VERSION_NUM >= 140000
#include "utils/backend_progress.h"
#else
#elif PG_VERSION_NUM >= 120000
#include "pgstat.h"
#endif
#if PG_VERSION_NUM >= 120000
#include "access/tableam.h"
#include "commands/progress.h"
#else
#define PROGRESS_CREATEIDX_SUBPHASE 0
#define PROGRESS_CREATEIDX_TUPLES_TOTAL 0
#define PROGRESS_CREATEIDX_TUPLES_DONE 0
#endif
#if PG_VERSION_NUM >= 130000
#define CALLBACK_ITEM_POINTER ItemPointer tid
#else
#define CALLBACK_ITEM_POINTER HeapTuple hup
#endif
#if PG_VERSION_NUM >= 120000
#define UpdateProgress(index, val) pgstat_progress_update_param(index, val)
#else
#define UpdateProgress(index, val) ((void)val)
#endif
#if PG_VERSION_NUM >= 140000
#include "utils/backend_status.h"
#include "utils/wait_event.h"
#endif
#if PG_VERSION_NUM >= 120000
#include "access/table.h"
#include "optimizer/optimizer.h"
#else
#include "access/heapam.h"
#include "optimizer/planner.h"
#include "pgstat.h"
#endif
#define PARALLEL_KEY_IVFFLAT_SHARED UINT64CONST(0xA000000000000001)
#define PARALLEL_KEY_TUPLESORT UINT64CONST(0xA000000000000002)
#define PARALLEL_KEY_IVFFLAT_CENTERS UINT64CONST(0xA000000000000003)
@@ -130,8 +150,13 @@ SampleRows(IvfflatBuildState * buildstate)
{
BlockNumber targblock = BlockSampler_Next(&buildstate->bs);
#if PG_VERSION_NUM >= 120000
table_index_build_range_scan(buildstate->heap, buildstate->index, buildstate->indexInfo,
false, true, false, targblock, 1, SampleCallback, (void *) buildstate, NULL);
#else
IndexBuildHeapRangeScan(buildstate->heap, buildstate->index, buildstate->indexInfo,
false, true, targblock, 1, SampleCallback, (void *) buildstate, NULL);
#endif
}
}
@@ -257,12 +282,16 @@ InsertTuples(Relation index, IvfflatBuildState * buildstate, ForkNumber forkNum)
IndexTuple itup = NULL; /* silence compiler warning */
int64 inserted = 0;
#if PG_VERSION_NUM >= 120000
TupleTableSlot *slot = MakeSingleTupleTableSlot(buildstate->tupdesc, &TTSOpsMinimalTuple);
#else
TupleTableSlot *slot = MakeSingleTupleTableSlot(buildstate->tupdesc);
#endif
TupleDesc tupdesc = RelationGetDescr(index);
pgstat_progress_update_param(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_IVFFLAT_PHASE_LOAD);
UpdateProgress(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_IVFFLAT_PHASE_LOAD);
pgstat_progress_update_param(PROGRESS_CREATEIDX_TUPLES_TOTAL, buildstate->indtuples);
UpdateProgress(PROGRESS_CREATEIDX_TUPLES_TOTAL, buildstate->indtuples);
GetNextTuple(buildstate->sortstate, tupdesc, slot, &itup, &list);
@@ -298,7 +327,7 @@ InsertTuples(Relation index, IvfflatBuildState * buildstate, ForkNumber forkNum)
pfree(itup);
pgstat_progress_update_param(PROGRESS_CREATEIDX_TUPLES_DONE, ++inserted);
UpdateProgress(PROGRESS_CREATEIDX_TUPLES_DONE, ++inserted);
GetNextTuple(buildstate->sortstate, tupdesc, slot, &itup, &list);
}
@@ -346,12 +375,20 @@ InitBuildState(IvfflatBuildState * buildstate, Relation heap, Relation index, In
elog(ERROR, "dimensions must be greater than one for this opclass");
/* Create tuple description for sorting */
#if PG_VERSION_NUM >= 120000
buildstate->tupdesc = CreateTemplateTupleDesc(3);
#else
buildstate->tupdesc = CreateTemplateTupleDesc(3, false);
#endif
TupleDescInitEntry(buildstate->tupdesc, (AttrNumber) 1, "list", INT4OID, -1, 0);
TupleDescInitEntry(buildstate->tupdesc, (AttrNumber) 2, "tid", TIDOID, -1, 0);
TupleDescInitEntry(buildstate->tupdesc, (AttrNumber) 3, "vector", RelationGetDescr(index)->attrs[0].atttypid, -1, 0);
#if PG_VERSION_NUM >= 120000
buildstate->slot = MakeSingleTupleTableSlot(buildstate->tupdesc, &TTSOpsVirtual);
#else
buildstate->slot = MakeSingleTupleTableSlot(buildstate->tupdesc);
#endif
buildstate->centers = VectorArrayInit(buildstate->lists, buildstate->dimensions);
buildstate->listInfo = palloc(sizeof(ListInfo) * buildstate->lists);
@@ -398,7 +435,7 @@ ComputeCenters(IvfflatBuildState * buildstate)
{
int numSamples;
pgstat_progress_update_param(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_IVFFLAT_PHASE_KMEANS);
UpdateProgress(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_IVFFLAT_PHASE_KMEANS);
/* Target 50 samples per list, with at least 10000 samples */
/* The number of samples has a large effect on index build time */
@@ -473,7 +510,7 @@ CreateListPages(Relation index, VectorArray centers, int dimensions,
IvfflatList list;
listSize = MAXALIGN(IVFFLAT_LIST_SIZE(dimensions));
list = palloc0(listSize);
list = palloc(listSize);
buf = IvfflatNewBuffer(index, forkNum);
IvfflatInitRegisterPage(index, &buf, &page, &state);
@@ -594,7 +631,11 @@ IvfflatParallelScanAndSort(IvfflatSpool * ivfspool, IvfflatShared * ivfshared, S
{
SortCoordinate coordinate;
IvfflatBuildState buildstate;
#if PG_VERSION_NUM >= 120000
TableScanDesc scan;
#else
HeapScanDesc scan;
#endif
double reltuples;
IndexInfo *indexInfo;
@@ -618,11 +659,18 @@ IvfflatParallelScanAndSort(IvfflatSpool * ivfspool, IvfflatShared * ivfshared, S
buildstate.centers->length = buildstate.centers->maxlen;
ivfspool->sortstate = tuplesort_begin_heap(buildstate.tupdesc, 1, attNums, sortOperators, sortCollations, nullsFirstFlags, sortmem, coordinate, false);
buildstate.sortstate = ivfspool->sortstate;
#if PG_VERSION_NUM >= 120000
scan = table_beginscan_parallel(ivfspool->heap,
ParallelTableScanFromIvfflatShared(ivfshared));
reltuples = table_index_build_scan(ivfspool->heap, ivfspool->index, indexInfo,
true, progress, BuildCallback,
(void *) &buildstate, scan);
#else
scan = heap_beginscan_parallel(ivfspool->heap, &ivfshared->heapdesc);
reltuples = IndexBuildHeapScan(ivfspool->heap, ivfspool->index, indexInfo,
true, BuildCallback,
(void *) &buildstate, scan);
#endif
/* Execute this worker's part of the sort */
tuplesort_performsort(ivfspool->sortstate);
@@ -692,7 +740,11 @@ IvfflatParallelBuildMain(dsm_segment *seg, shm_toc *toc)
}
/* Open relations within worker */
#if PG_VERSION_NUM >= 120000
heapRel = table_open(ivfshared->heaprelid, heapLockmode);
#else
heapRel = heap_open(ivfshared->heaprelid, heapLockmode);
#endif
indexRel = index_open(ivfshared->indexrelid, indexLockmode);
/* Initialize worker's own spool */
@@ -712,7 +764,11 @@ IvfflatParallelBuildMain(dsm_segment *seg, shm_toc *toc)
/* Close relations within worker */
index_close(indexRel, indexLockmode);
#if PG_VERSION_NUM >= 120000
table_close(heapRel, heapLockmode);
#else
heap_close(heapRel, heapLockmode);
#endif
}
/*
@@ -737,7 +793,19 @@ IvfflatEndParallel(IvfflatLeader * ivfleader)
static Size
ParallelEstimateShared(Relation heap, Snapshot snapshot)
{
#if PG_VERSION_NUM >= 120000
return add_size(BUFFERALIGN(sizeof(IvfflatShared)), table_parallelscan_estimate(heap, snapshot));
#else
if (!IsMVCCSnapshot(snapshot))
{
Assert(snapshot == SnapshotAny);
return sizeof(IvfflatShared);
}
return add_size(offsetof(IvfflatShared, heapdesc) +
offsetof(ParallelHeapScanDescData, phs_snapshot_data),
EstimateSnapshotSpace(snapshot));
#endif
}
/*
@@ -788,7 +856,11 @@ IvfflatBeginParallel(IvfflatBuildState * buildstate, bool isconcurrent, int requ
/* Enter parallel mode and create context */
EnterParallelMode();
Assert(request > 0);
#if PG_VERSION_NUM >= 120000
pcxt = CreateParallelContext("vector", "IvfflatParallelBuildMain", request);
#else
pcxt = CreateParallelContext("vector", "IvfflatParallelBuildMain", request, true);
#endif
scantuplesortstates = leaderparticipates ? request + 1 : request;
@@ -846,9 +918,13 @@ IvfflatBeginParallel(IvfflatBuildState * buildstate, bool isconcurrent, int requ
#ifdef IVFFLAT_KMEANS_DEBUG
ivfshared->inertia = 0;
#endif
#if PG_VERSION_NUM >= 120000
table_parallelscan_initialize(buildstate->heap,
ParallelTableScanFromIvfflatShared(ivfshared),
snapshot);
#else
heap_parallelscan_initialize(&ivfshared->heapdesc, buildstate->heap, snapshot);
#endif
/* Store shared tuplesort-private state, for which we reserved space */
sharedsort = (Sharedsort *) shm_toc_allocate(pcxt->toc, estsort);
@@ -919,7 +995,7 @@ AssignTuples(IvfflatBuildState * buildstate)
Oid sortCollations[] = {InvalidOid};
bool nullsFirstFlags[] = {false};
pgstat_progress_update_param(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_IVFFLAT_PHASE_ASSIGN);
UpdateProgress(PROGRESS_CREATEIDX_SUBPHASE, PROGRESS_IVFFLAT_PHASE_ASSIGN);
/* Calculate parallel workers */
if (buildstate->heap != NULL)
@@ -947,8 +1023,15 @@ AssignTuples(IvfflatBuildState * buildstate)
if (buildstate->ivfleader)
buildstate->reltuples = ParallelHeapScan(buildstate);
else
{
#if PG_VERSION_NUM >= 120000
buildstate->reltuples = table_index_build_scan(buildstate->heap, buildstate->index, buildstate->indexInfo,
true, true, BuildCallback, (void *) buildstate, NULL);
#else
buildstate->reltuples = IndexBuildHeapScan(buildstate->heap, buildstate->index, buildstate->indexInfo,
true, BuildCallback, (void *) buildstate, NULL);
#endif
}
#ifdef IVFFLAT_KMEANS_DEBUG
PrintKmeansMetrics(buildstate);

View File

@@ -3,16 +3,14 @@
#include <float.h>
#include "access/amapi.h"
#include "access/reloptions.h"
#include "commands/progress.h"
#include "commands/vacuum.h"
#include "ivfflat.h"
#include "utils/guc.h"
#include "utils/selfuncs.h"
#include "utils/spccache.h"
#if PG_VERSION_NUM < 150000
#define MarkGUCPrefixReserved(x) EmitWarningsOnPlaceholders(x)
#if PG_VERSION_NUM >= 120000
#include "commands/progress.h"
#endif
int ivfflat_probes;
@@ -35,13 +33,12 @@ IvfflatInit(void)
DefineCustomIntVariable("ivfflat.probes", "Sets the number of probes",
"Valid range is 1..lists.", &ivfflat_probes,
IVFFLAT_DEFAULT_PROBES, IVFFLAT_MIN_LISTS, IVFFLAT_MAX_LISTS, PGC_USERSET, 0, NULL, NULL, NULL);
MarkGUCPrefixReserved("ivfflat");
}
/*
* Get the name of index build phase
*/
#if PG_VERSION_NUM >= 120000
static char *
ivfflatbuildphasename(int64 phasenum)
{
@@ -59,6 +56,7 @@ ivfflatbuildphasename(int64 phasenum)
return NULL;
}
}
#endif
/*
* Estimate the cost of an index scan
@@ -74,6 +72,9 @@ ivfflatcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
double ratio;
double spc_seq_page_cost;
Relation index;
#if PG_VERSION_NUM < 120000
List *qinfos;
#endif
/* Never use index without order */
if (path->indexorderbys == NULL)
@@ -104,7 +105,26 @@ ivfflatcostestimate(PlannerInfo *root, IndexPath *path, double loop_count,
*/
costs.numIndexTuples = path->indexinfo->tuples * ratio;
/*
* Do not use index if no limit or limit + offset > expected tuples unless
* enable_seqscan = off
*/
if (root->limit_tuples < 0 || root->limit_tuples > costs.numIndexTuples)
{
*indexStartupCost = 1.0e10 - 1;
*indexTotalCost = 1.0e10 - 1;
*indexSelectivity = 0;
*indexCorrelation = 0;
*indexPages = 0;
return;
}
#if PG_VERSION_NUM >= 120000
genericcostestimate(root, path, loop_count, &costs);
#else
qinfos = deconstruct_indexquals(path);
genericcostestimate(root, path, loop_count, qinfos, &costs);
#endif
get_tablespace_page_costs(path->indexinfo->reltablespace, NULL, &spc_seq_page_cost);
@@ -221,7 +241,9 @@ ivfflathandler(PG_FUNCTION_ARGS)
amroutine->amcostestimate = ivfflatcostestimate;
amroutine->amoptions = ivfflatoptions;
amroutine->amproperty = NULL; /* TODO AMPROP_DISTANCE_ORDERABLE */
#if PG_VERSION_NUM >= 120000
amroutine->ambuildphasename = ivfflatbuildphasename;
#endif
amroutine->amvalidate = ivfflatvalidate;
#if PG_VERSION_NUM >= 140000
amroutine->amadjustmembers = NULL;

View File

@@ -3,10 +3,9 @@
#include "postgres.h"
#include "access/genam.h"
#include "access/generic_xlog.h"
#include "access/parallel.h"
#include "lib/pairingheap.h"
#include "access/reloptions.h"
#include "nodes/execnodes.h"
#include "port.h" /* for random() */
#include "utils/sampling.h"
@@ -17,6 +16,10 @@
#include "common/pg_prng.h"
#endif
#if PG_VERSION_NUM < 120000
#include "access/relscan.h"
#endif
#ifdef IVFFLAT_BENCH
#include "portability/instr_time.h"
#endif
@@ -132,10 +135,16 @@ typedef struct IvfflatShared
#ifdef IVFFLAT_KMEANS_DEBUG
double inertia;
#endif
#if PG_VERSION_NUM < 120000
ParallelHeapScanDescData heapdesc; /* must come last */
#endif
} IvfflatShared;
#if PG_VERSION_NUM >= 120000
#define ParallelTableScanFromIvfflatShared(shared) \
(ParallelTableScanDesc) ((char *) (shared) + BUFFERALIGN(sizeof(IvfflatShared)))
#endif
typedef struct IvfflatLeader
{

View File

@@ -2,7 +2,6 @@
#include <float.h>
#include "access/generic_xlog.h"
#include "ivfflat.h"
#include "storage/bufmgr.h"
#include "storage/lmgr.h"

View File

@@ -5,7 +5,6 @@
#include "access/relscan.h"
#include "catalog/pg_operator_d.h"
#include "catalog/pg_type_d.h"
#include "lib/pairingheap.h"
#include "ivfflat.h"
#include "miscadmin.h"
#include "pgstat.h"
@@ -106,7 +105,12 @@ GetScanItems(IndexScanDesc scan, Datum value)
IvfflatScanOpaque so = (IvfflatScanOpaque) scan->opaque;
TupleDesc tupdesc = RelationGetDescr(scan->indexRelation);
double tuples = 0;
#if PG_VERSION_NUM >= 120000
TupleTableSlot *slot = MakeSingleTupleTableSlot(so->tupdesc, &TTSOpsVirtual);
#else
TupleTableSlot *slot = MakeSingleTupleTableSlot(so->tupdesc);
#endif
/*
* Reuse same set of shared buffers for scan
@@ -212,14 +216,22 @@ ivfflatbeginscan(Relation index, int nkeys, int norderbys)
so->collation = index->rd_indcollation[0];
/* Create tuple description for sorting */
#if PG_VERSION_NUM >= 120000
so->tupdesc = CreateTemplateTupleDesc(2);
#else
so->tupdesc = CreateTemplateTupleDesc(2, false);
#endif
TupleDescInitEntry(so->tupdesc, (AttrNumber) 1, "distance", FLOAT8OID, -1, 0);
TupleDescInitEntry(so->tupdesc, (AttrNumber) 2, "heaptid", TIDOID, -1, 0);
/* Prep sort */
so->sortstate = tuplesort_begin_heap(so->tupdesc, 1, attNums, sortOperators, sortCollations, nullsFirstFlags, work_mem, NULL, false);
#if PG_VERSION_NUM >= 120000
so->slot = MakeSingleTupleTableSlot(so->tupdesc, &TTSOpsMinimalTuple);
#else
so->slot = MakeSingleTupleTableSlot(so->tupdesc);
#endif
so->listQueue = pairingheap_allocate(CompareLists, scan);
@@ -309,8 +321,12 @@ ivfflatgettuple(IndexScanDesc scan, ScanDirection dir)
{
ItemPointer heaptid = (ItemPointer) DatumGetPointer(slot_getattr(so->slot, 2, &so->isnull));
#if PG_VERSION_NUM >= 120000
scan->xs_heaptid = *heaptid;
scan->xs_recheck = false;
#else
scan->xs_ctup.t_self = *heaptid;
#endif
scan->xs_recheckorderby = false;
return true;
}

View File

@@ -1,6 +1,5 @@
#include "postgres.h"
#include "access/generic_xlog.h"
#include "ivfflat.h"
#include "storage/bufmgr.h"
#include "vector.h"

View File

@@ -1,6 +1,5 @@
#include "postgres.h"
#include "access/generic_xlog.h"
#include "commands/vacuum.h"
#include "ivfflat.h"
#include "storage/bufmgr.h"

View File

@@ -3,7 +3,6 @@
#include <math.h>
#include "catalog/pg_type.h"
#include "common/shortest_dec.h"
#include "fmgr.h"
#include "hnsw.h"
#include "ivfflat.h"
@@ -12,7 +11,6 @@
#include "port.h" /* for strtof() */
#include "utils/array.h"
#include "utils/builtins.h"
#include "utils/float.h"
#include "utils/lsyscache.h"
#include "utils/numeric.h"
#include "vector.h"
@@ -21,6 +19,13 @@
#include "varatt.h"
#endif
#if PG_VERSION_NUM >= 120000
#include "common/shortest_dec.h"
#include "utils/float.h"
#else
#include <float.h>
#endif
#if PG_VERSION_NUM < 130000
#define TYPALIGN_DOUBLE 'd'
#define TYPALIGN_INT 'i'
@@ -288,6 +293,15 @@ vector_out(PG_FUNCTION_ARGS)
char *ptr;
int n;
#if PG_VERSION_NUM < 120000
int ndig = FLT_DIG + extra_float_digits;
if (ndig < 1)
ndig = 1;
#define FLOAT_SHORTEST_DECIMAL_LEN (ndig + 10)
#endif
/*
* Need:
*
@@ -311,7 +325,11 @@ vector_out(PG_FUNCTION_ARGS)
ptr++;
}
#if PG_VERSION_NUM >= 120000
n = float_to_shortest_decimal_bufn(vector->x[i], ptr);
#else
n = sprintf(ptr, "%.*g", ndig, vector->x[i]);
#endif
ptr += n;
}
*ptr = ']';
@@ -866,10 +884,9 @@ vector_mul(PG_FUNCTION_ARGS)
int
vector_cmp_internal(Vector * a, Vector * b)
{
int dim = Min(a->dim, b->dim);
CheckDims(a, b);
/* Check values before dimensions to be consistent with Postgres arrays */
for (int i = 0; i < dim; i++)
for (int i = 0; i < a->dim; i++)
{
if (a->x[i] < b->x[i])
return -1;
@@ -877,13 +894,6 @@ vector_cmp_internal(Vector * a, Vector * b)
if (a->x[i] > b->x[i])
return 1;
}
if (a->dim < b->dim)
return -1;
if (a->dim > b->dim)
return 1;
return 0;
}
@@ -897,9 +907,6 @@ vector_lt(PG_FUNCTION_ARGS)
Vector *a = PG_GETARG_VECTOR_P(0);
Vector *b = PG_GETARG_VECTOR_P(1);
/* TODO Remove in 0.7.0 */
CheckDims(a, b);
PG_RETURN_BOOL(vector_cmp_internal(a, b) < 0);
}
@@ -913,9 +920,6 @@ vector_le(PG_FUNCTION_ARGS)
Vector *a = PG_GETARG_VECTOR_P(0);
Vector *b = PG_GETARG_VECTOR_P(1);
/* TODO Remove in 0.7.0 */
CheckDims(a, b);
PG_RETURN_BOOL(vector_cmp_internal(a, b) <= 0);
}
@@ -929,9 +933,6 @@ vector_eq(PG_FUNCTION_ARGS)
Vector *a = PG_GETARG_VECTOR_P(0);
Vector *b = PG_GETARG_VECTOR_P(1);
/* TODO Remove in 0.7.0 */
CheckDims(a, b);
PG_RETURN_BOOL(vector_cmp_internal(a, b) == 0);
}
@@ -945,9 +946,6 @@ vector_ne(PG_FUNCTION_ARGS)
Vector *a = PG_GETARG_VECTOR_P(0);
Vector *b = PG_GETARG_VECTOR_P(1);
/* TODO Remove in 0.7.0 */
CheckDims(a, b);
PG_RETURN_BOOL(vector_cmp_internal(a, b) != 0);
}
@@ -961,9 +959,6 @@ vector_ge(PG_FUNCTION_ARGS)
Vector *a = PG_GETARG_VECTOR_P(0);
Vector *b = PG_GETARG_VECTOR_P(1);
/* TODO Remove in 0.7.0 */
CheckDims(a, b);
PG_RETURN_BOOL(vector_cmp_internal(a, b) >= 0);
}
@@ -977,9 +972,6 @@ vector_gt(PG_FUNCTION_ARGS)
Vector *a = PG_GETARG_VECTOR_P(0);
Vector *b = PG_GETARG_VECTOR_P(1);
/* TODO Remove in 0.7.0 */
CheckDims(a, b);
PG_RETURN_BOOL(vector_cmp_internal(a, b) > 0);
}

View File

@@ -24,56 +24,6 @@ SELECT '[1e37]'::vector * '[1e37]';
ERROR: value out of range: overflow
SELECT '[1e-37]'::vector * '[1e-37]';
ERROR: value out of range: underflow
SELECT '[1,2,3]'::vector = '[1,2,3]';
?column?
----------
t
(1 row)
SELECT '[1,2,3]'::vector = '[1,2]';
ERROR: different vector dimensions 3 and 2
SELECT vector_cmp('[1,2,3]', '[1,2,3]');
vector_cmp
------------
0
(1 row)
SELECT vector_cmp('[1,2,3]', '[0,0,0]');
vector_cmp
------------
1
(1 row)
SELECT vector_cmp('[0,0,0]', '[1,2,3]');
vector_cmp
------------
-1
(1 row)
SELECT vector_cmp('[1,2]', '[1,2,3]');
vector_cmp
------------
-1
(1 row)
SELECT vector_cmp('[1,2,3]', '[1,2]');
vector_cmp
------------
1
(1 row)
SELECT vector_cmp('[1,2]', '[2,3,4]');
vector_cmp
------------
-1
(1 row)
SELECT vector_cmp('[2,3]', '[1,2,3]');
vector_cmp
------------
1
(1 row)
SELECT vector_dims('[1,2,3]');
vector_dims
-------------

View File

@@ -6,17 +6,6 @@ SELECT '[1,2,3]'::vector * '[4,5,6]';
SELECT '[1e37]'::vector * '[1e37]';
SELECT '[1e-37]'::vector * '[1e-37]';
SELECT '[1,2,3]'::vector = '[1,2,3]';
SELECT '[1,2,3]'::vector = '[1,2]';
SELECT vector_cmp('[1,2,3]', '[1,2,3]');
SELECT vector_cmp('[1,2,3]', '[0,0,0]');
SELECT vector_cmp('[0,0,0]', '[1,2,3]');
SELECT vector_cmp('[1,2]', '[1,2,3]');
SELECT vector_cmp('[1,2,3]', '[1,2]');
SELECT vector_cmp('[1,2]', '[2,3,4]');
SELECT vector_cmp('[2,3]', '[1,2,3]');
SELECT vector_dims('[1,2,3]');
SELECT round(vector_norm('[1,1]')::numeric, 5);

View File

@@ -95,10 +95,11 @@ for my $i (0 .. $#operators)
$node->safe_psql("postgres", "DROP INDEX idx;");
# Build index in parallel in memory
# Build index in parallel
my ($ret, $stdout, $stderr) = $node->psql("postgres", qq(
SET client_min_messages = DEBUG;
SET min_parallel_table_scan_size = 1;
SET hnsw.enable_parallel_build = on;
CREATE INDEX idx ON tst USING hnsw (v $opclass);
));
is($ret, 0, $stderr);
@@ -108,21 +109,6 @@ for my $i (0 .. $#operators)
test_recall($min, $operator);
$node->safe_psql("postgres", "DROP INDEX idx;");
# Build index in parallel on disk
# Set parallel_workers on table to use workers with low maintenance_work_mem
($ret, $stdout, $stderr) = $node->psql("postgres", qq(
ALTER TABLE tst SET (parallel_workers = 2);
SET client_min_messages = DEBUG;
SET maintenance_work_mem = '4MB';
CREATE INDEX idx ON tst USING hnsw (v $opclass);
ALTER TABLE tst RESET (parallel_workers);
));
is($ret, 0, $stderr);
like($stderr, qr/using \d+ parallel workers/);
like($stderr, qr/hnsw graph no longer fits into maintenance_work_mem/);
$node->safe_psql("postgres", "DROP INDEX idx;");
}
done_testing();

View File

@@ -1,114 +0,0 @@
use strict;
use warnings;
use PostgresNode;
use TestLib;
use Test::More;
my $dim = 3;
my $nc = 50;
my $limit = 20;
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;");
$node->safe_psql("postgres", "CREATE TABLE tst (i int4, v vector($dim), c int4, t text);");
$node->safe_psql("postgres",
"INSERT INTO tst SELECT i, ARRAY[$array_sql], i % $nc, 'test ' || i FROM generate_series(1, 10000) i;"
);
$node->safe_psql("postgres", "CREATE INDEX idx ON tst USING hnsw (v vector_l2_ops);");
$node->safe_psql("postgres", "ANALYZE tst;");
# Generate query
my @r = ();
for (1 .. $dim)
{
push(@r, rand());
}
my $query = "[" . join(",", @r) . "]";
my $c = int(rand() * $nc);
# Test attribute filtering
my $explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE c = $c ORDER BY v <-> '$query' LIMIT $limit;
));
# TODO Do not use index
like($explain, qr/Index Scan using idx/);
# Test attribute filtering with few rows removed
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE c != $c ORDER BY v <-> '$query' LIMIT $limit;
));
like($explain, qr/Index Scan using idx/);
# Test attribute filtering with few rows removed comparison
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE c >= 1 ORDER BY v <-> '$query' LIMIT $limit;
));
like($explain, qr/Index Scan using idx/);
# Test attribute filtering with many rows removed comparison
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE c < 1 ORDER BY v <-> '$query' LIMIT $limit;
));
# TODO Do not use index
like($explain, qr/Index Scan using idx/);
# Test attribute filtering with few rows removed like
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE t LIKE '%%test%%' ORDER BY v <-> '$query' LIMIT $limit;
));
like($explain, qr/Index Scan using idx/);
# Test attribute filtering with many rows removed like
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE t LIKE '%%other%%' ORDER BY v <-> '$query' LIMIT $limit;
));
like($explain, qr/Seq Scan/);
# Test distance filtering
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE v <-> '$query' < 1 ORDER BY v <-> '$query' LIMIT $limit;
));
like($explain, qr/Index Scan using idx/);
# Test distance filtering greater than distance
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE v <-> '$query' > 1 ORDER BY v <-> '$query' LIMIT $limit;
));
# TODO Do not use index
like($explain, qr/Index Scan using idx/);
# Test distance filtering without order
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE v <-> '$query' < 1;
));
like($explain, qr/Seq Scan/);
# Test distance filtering without limit
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE v <-> '$query' < 1 ORDER BY v <-> '$query';
));
like($explain, qr/Seq Scan/);
# Test attribute index
$node->safe_psql("postgres", "CREATE INDEX attribute_idx ON tst (c);");
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE c = $c ORDER BY v <-> '$query' LIMIT $limit;
));
# TODO Use attribute index
like($explain, qr/Index Scan using idx/);
# Test partial index
$node->safe_psql("postgres", "CREATE INDEX partial_idx ON tst USING hnsw (v vector_l2_ops) WHERE (c = $c);");
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE c = $c ORDER BY v <-> '$query' LIMIT $limit;
));
like($explain, qr/Index Scan using partial_idx/);
done_testing();

View File

@@ -1,116 +0,0 @@
use strict;
use warnings;
use PostgresNode;
use TestLib;
use Test::More;
my $dim = 3;
my $nc = 50;
my $limit = 20;
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;");
$node->safe_psql("postgres", "CREATE TABLE tst (i int4, v vector($dim), c int4, t text);");
$node->safe_psql("postgres",
"INSERT INTO tst SELECT i, ARRAY[$array_sql], i % $nc, 'test ' || i FROM generate_series(1, 10000) i;"
);
$node->safe_psql("postgres", "CREATE INDEX idx ON tst USING ivfflat (v vector_l2_ops) WITH (lists = 100);");
$node->safe_psql("postgres", "ANALYZE tst;");
# Generate query
my @r = ();
for (1 .. $dim)
{
push(@r, rand());
}
my $query = "[" . join(",", @r) . "]";
my $c = int(rand() * $nc);
# Test attribute filtering
my $explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE c = $c ORDER BY v <-> '$query' LIMIT $limit;
));
# TODO Do not use index
like($explain, qr/Index Scan using idx/);
# Test attribute filtering with few rows removed
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE c != $c ORDER BY v <-> '$query' LIMIT $limit;
));
like($explain, qr/Index Scan using idx/);
# Test attribute filtering with few rows removed comparison
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE c >= 1 ORDER BY v <-> '$query' LIMIT $limit;
));
like($explain, qr/Index Scan using idx/);
# Test attribute filtering with many rows removed comparison
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE c < 1 ORDER BY v <-> '$query' LIMIT $limit;
));
# TODO Do not use index
like($explain, qr/Index Scan using idx/);
# Test attribute filtering with few rows removed like
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE t LIKE '%%test%%' ORDER BY v <-> '$query' LIMIT $limit;
));
like($explain, qr/Index Scan using idx/);
# Test attribute filtering with many rows removed like
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE t LIKE '%%other%%' ORDER BY v <-> '$query' LIMIT $limit;
));
like($explain, qr/Seq Scan/);
# Test distance filtering
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE v <-> '$query' < 1 ORDER BY v <-> '$query' LIMIT $limit;
));
like($explain, qr/Index Scan using idx/);
# Test distance filtering greater than distance
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE v <-> '$query' > 1 ORDER BY v <-> '$query' LIMIT $limit;
));
# TODO Do not use index
like($explain, qr/Index Scan using idx/);
# Test distance filtering without order
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE v <-> '$query' < 1;
));
like($explain, qr/Seq Scan/);
# Test distance filtering without limit
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE v <-> '$query' < 1 ORDER BY v <-> '$query';
));
# TODO Do not use index
like($explain, qr/Index Scan using idx/);
# Test attribute index
$node->safe_psql("postgres", "CREATE INDEX attribute_idx ON tst (c);");
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE c = $c ORDER BY v <-> '$query' LIMIT $limit;
));
# TODO Use attribute index
like($explain, qr/Index Scan using idx/);
# Test partial index
$node->safe_psql("postgres", "CREATE INDEX partial_idx ON tst USING ivfflat (v vector_l2_ops) WITH (lists = 5) WHERE (c = $c);");
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT i FROM tst WHERE c = $c ORDER BY v <-> '$query' LIMIT $limit;
));
# TODO Use partial index
like($explain, qr/Index Scan using idx/);
done_testing();

View File

@@ -0,0 +1,64 @@
use strict;
use warnings;
use PostgresNode;
use TestLib;
use Test::More;
# 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 (v vector(3));");
$node->safe_psql("postgres",
"INSERT INTO tst SELECT ARRAY[random(), random(), random()] FROM generate_series(1, 1000) i;"
);
$node->safe_psql("postgres", "CREATE INDEX ON tst USING ivfflat (v vector_l2_ops) WITH (lists = 10);");
# Test limit
my $explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT * FROM tst ORDER BY v <-> '[1,2,3]' LIMIT 100;
));
like($explain, qr/Index Scan/);
# Test limit with probes
$explain = $node->safe_psql("postgres", qq(
SET ivfflat.probes = 2;
EXPLAIN ANALYZE SELECT * FROM tst ORDER BY v <-> '[1,2,3]' LIMIT 200;
));
like($explain, qr/Index Scan/);
# Test limit + offset
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT * FROM tst ORDER BY v <-> '[1,2,3]' LIMIT 90 OFFSET 10;
));
like($explain, qr/Index Scan/);
# Test limit > expected tuples
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT * FROM tst ORDER BY v <-> '[1,2,3]' LIMIT 101;
));
like($explain, qr/Seq Scan/);
# Test limit > expected tuples with probes
$explain = $node->safe_psql("postgres", qq(
SET ivfflat.probes = 2;
EXPLAIN ANALYZE SELECT * FROM tst ORDER BY v <-> '[1,2,3]' LIMIT 201;
));
like($explain, qr/Seq Scan/);
# Test limit + offset > expected tuples
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT * FROM tst ORDER BY v <-> '[1,2,3]' LIMIT 91 OFFSET 10;
));
like($explain, qr/Seq Scan/);
# Test no limit
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT * FROM tst ORDER BY v <-> '[1,2,3]';
));
like($explain, qr/Seq Scan/);
done_testing();

View File

@@ -1,113 +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;"
);
$node->safe_psql("postgres", qq(
CREATE FUNCTION float4_l2_distance(float4[], float4[]) RETURNS float8
AS 'BEGIN RETURN l2_distance(\$1::vector, \$2::vector); END;'
LANGUAGE plpgsql IMMUTABLE STRICT PARALLEL SAFE;
CREATE FUNCTION float4_l2_squared_distance(float4[], float4[]) RETURNS float8
AS 'BEGIN RETURN vector_l2_squared_distance(\$1::vector, \$2::vector); END;'
LANGUAGE plpgsql IMMUTABLE STRICT PARALLEL SAFE;
CREATE OPERATOR <-> (
LEFTARG = float4[], RIGHTARG = float4[], PROCEDURE = float4_l2_distance,
COMMUTATOR = '<->'
);
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[]);
));
# 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);");
my $min = $operator eq "<#>" ? 0.80 : 0.99;
test_recall($min, $operator);
}
done_testing();

62
test/t/020_hnsw_limit.pl Normal file
View File

@@ -0,0 +1,62 @@
use strict;
use warnings;
use PostgresNode;
use TestLib;
use Test::More;
# 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 (v vector(3));");
$node->safe_psql("postgres",
"INSERT INTO tst SELECT ARRAY[random(), random(), random()] FROM generate_series(1, 1000) i;"
);
$node->safe_psql("postgres", "CREATE INDEX ON tst USING hnsw (v vector_l2_ops);");
# Test limit
my $explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT * FROM tst ORDER BY v <-> '[1,2,3]' LIMIT 40;
));
like($explain, qr/Index Scan/);
# Test limit with CTE
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE WITH cte AS (SELECT * FROM tst ORDER BY v <-> '[1,2,3]' LIMIT 40) SELECT * FROM cte;
));
like($explain, qr/Index Scan/);
# Test limit + offset
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT * FROM tst ORDER BY v <-> '[1,2,3]' LIMIT 30 OFFSET 10;
));
like($explain, qr/Index Scan/);
# Test limit > ef_search
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT * FROM tst ORDER BY v <-> '[1,2,3]' LIMIT 41;
));
like($explain, qr/Seq Scan/);
# Test limit > ef_search with CTE
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE WITH cte AS (SELECT * FROM tst ORDER BY v <-> '[1,2,3]' LIMIT 41) SELECT * FROM cte;
));
like($explain, qr/Seq Scan/);
# Test limit + offset > ef_search
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT * FROM tst ORDER BY v <-> '[1,2,3]' LIMIT 31 OFFSET 10;
));
like($explain, qr/Seq Scan/);
# Test no limit
$explain = $node->safe_psql("postgres", qq(
EXPLAIN ANALYZE SELECT * FROM tst ORDER BY v <-> '[1,2,3]';
));
like($explain, qr/Seq Scan/);
done_testing();

View File

@@ -1,4 +1,4 @@
comment = 'vector data type and ivfflat and hnsw access methods'
default_version = '0.6.0'
default_version = '0.5.1'
module_pathname = '$libdir/vector'
relocatable = true