openGauss内核分析（十）：数据库索引的创建过程

joytrian

发布于 2023-5-5 11:00

浏览

0收藏

数据库索引可以提高数据的访问速度，openGauss支持唯一索引、多字段索引、部分索引和表达式索引。行存表（ASTORE存储引擎）支持的索引类型：btree（行存表缺省值）、hash、gin、gist。行存表（USTORE存储引擎）支持的索引类型：ubtree。列存表支持的索引类型：Psort（列存表缺省值）、btree、gin。全局临时表不支持GIN索引和Gist索引。

如上一篇分析数据库表的创建过程，standard_ProcessUtility函数会根据nodeTag(parsetree)的值来确定sql的操作类型，create table一般都是进入T_CreateStmt分支调用CreateCommand函数。create index则进入T_IndexStmt分支调用DefineIndex函数。在调用DefineIndex前会首先执行函数transformIndexStmt，如果语句没有指定索引类型则会使用缺省值。

 if (stmt->accessMethod == NULL) {
        if (!isColStore) {
            /* row store using btree index by default */
            if (!RelationIsUstoreFormat(rel)) {
                stmt->accessMethod = DEFAULT_INDEX_TYPE; //  行存表ASTORE存储引擎缺省值btree
            } else {
                stmt->accessMethod = DEFAULT_USTORE_INDEX_TYPE; // 行存表USTORE存储引擎缺省值ubtree
            }
……
            } else {
                /* column store using psort index by default */
                stmt->accessMethod = DEFAULT_CSTORE_INDEX_TYPE; // 列存表缺省值psort
            }
        }1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.

#define DEFAULT_INDEX_TYPE  "btree"
#define DEFAULT_HASH_INDEX_TYPE "hash"
#define DEFAULT_CSTORE_INDEX_TYPE "psort"
#define DEFAULT_GIST_INDEX_TYPE  "gist"
#define CSTORE_BTREE_INDEX_TYPE "cbtree"
#define DEFAULT_GIN_INDEX_TYPE "gin"
#define CSTORE_GINBTREE_INDEX_TYPE "cgin"
#define DEFAULT_USTORE_INDEX_TYPE "ubtree"1.
2.
3.
4.
5.
6.
7.
8.

普通表索引

DefineInde

DefineIndex为创建索引主入口函数。通常创建索引以Share锁锁定表，允许并发查询，但禁上对表进行修改。如果创建索引时指定关键字CONCURRENTLY以不阻塞DML的方式创建索引，即允许读取和更新表，以ShareUpdateExclusiveLock锁锁定表。

openGauss内核分析（十）：数据库索引的创建过程-鸿蒙开发者社区

lockmode = concurrent ? ShareUpdateExclusiveLock : ShareLock;
rel = heap_open(relationId, lockmode);1.
2.

如果没有指定索引名，ChooseIndexName根据规则生成索引名：

    /*
     * Select name for index if caller didn't specify.
     */
    indexRelationName = stmt->idxname;
    if (indexRelationName == NULL) {
        indexRelationName = ChooseIndexName(RelationGetRelationName(rel),
            namespaceId,
            indexColNames,
            stmt->excludeOpNames,
            stmt->primary,
            stmt->isconstraint);1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.

为index_create函数构造参数IndexInfo结构体：

    /*
     * Prepare arguments for index_create, primarily an IndexInfo structure.
     * Note that ii_Predicate must be in implicit-AND format.
     */
    indexInfo = makeNode(IndexInfo);1.
2.
3.
4.
5.

Index_create函数创建索引：

    /*
     * Make the catalog entries for the index, including constraints. Then, if
     * not skip_build || concurrent, actually build the index.
     */
indexRelationId = index_create(rel,
……1.
2.
3.
4.
5.
6.

关闭表并返回索引表id：

    heap_close(rel, NoLock);
    return indexRelationId;1.
2.

Index_create函数

打开系统表pg_class：

    pg_class = heap_open(RelationRelationId, RowExclusiveLock);1.

heap_create创建relcache和索引物理文件：

    /*
     * create the index relation's relcache entry and physical disk file. (If
     * we fail further down, it's the smgr's responsibility to remove the disk
     * file again.)
     */
    StorageType storage_type = RelationGetStorageType(heapRelation);
    indexRelation = heap_create(indexRelationName, namespaceId, tableSpaceId, indexRelationId, relFileNode,
        RELATION_CREATE_BUCKET(heapRelation) ? heapRelation->rd_bucketoid : InvalidOid, indexTupDesc, relKind,
        relpersistence, isLocalPart, false, shared_relation, mapped_relation, allow_system_table_mods,
        REL_CMPRS_NOT_SUPPORT, (Datum)reloptions, heapRelation->rd_rel->relowner, skip_create_storage,
        isUstore ? TAM_USTORE : TAM_HEAP, /* XXX: Index tables are by default HEAP Table Type */
        relindexsplit, storage_type, extra->crossBucket, accessMethodObjectId);1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.

将索引表元信息存入系统表pg_class：

    /*
     * store index's pg_class entry
     */
    InsertPgClassTuple(
        pg_class, indexRelation, RelationGetRelid(indexRelation), (Datum)0, reloptions, relKind, NULL);
 
    /* done with pg_class */
    heap_close(pg_class, RowExclusiveLock);1.
2.
3.
4.
5.
6.
7.
8.

将索引表元信息存入系统表pg_index：

     UpdateIndexRelation(indexRelationId,
        heapRelationId,
        indexInfo,
……1.
2.
3.
4.

Index_build建立索引：

    } else if (extra && (!extra->isPartitionedIndex || extra->isGlobalPartitionedIndex)) {
        /* support regular index or GLOBAL partition index */
        index_build(heapRelation, NULL, indexRelation, NULL, indexInfo, isprimary, false, PARTITION_TYPE(extra));
    }1.
2.
3.
4.

index_build

index_build调用index_build_storage，如果创建的是btree索引最终调用btbuild，如果是hash索引最终调用hashbuild，如果是psort则最终调用psortbuild，更多索引访问方法的信息可查看系统表pg_am。

        stats = index_build_storage(targetHeapRelation, targetIndexRelation, indexInfo);1.

static IndexBuildResult* index_build_storage(Relation heapRelation, Relation indexRelation, IndexInfo* indexInfo)
{
    RegProcedure procedure = indexRelation->rd_am->ambuild;
    Assert(RegProcedureIsValid(procedure));
 
    IndexBuildResult* stats = (IndexBuildResult*)DatumGetPointer(OidFunctionCall3(
        procedure, PointerGetDatum(heapRelation), PointerGetDatum(indexRelation), PointerGetDatum(indexInfo)));
    Assert(PointerIsValid(stats));
    if (RELPERSISTENCE_UNLOGGED == heapRelation->rd_rel->relpersistence) {
        index_build_init_fork(heapRelation, indexRelation);
    }
    return stats;
}1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.

btree索引的procedure为btbuild。

#0  btbuild (fcinfo=0x7fb4f9c63920) at nbtree.cpp:63
#1  0x00000000011fc07d in OidFunctionCall3Coll (functionId=338, collation=0, arg1=140415366247440, arg2=140415366237480, arg3=140415402419864) at fmgr.cpp:1857
#2  0x0000000000c16b77 in index_build_storage (heapRelation=0x7fb50006b410, indexRelation=0x7fb500068d28, indexInfo=0x7fb5022ea698) at index.cpp:2475
#3  0x0000000000c18097 in index_build (heapRelation=0x7fb50006b410, heapPartition=0x0, indexRelation=0x7fb500068d28, indexPartition=0x0, indexInfo=0x7fb5022ea698, isPrimary=false, isreindex=false,
    partitionType=INDEX_CREATE_NONE_PARTITION, parallel=true) at index.cpp:28341.
2.
3.
4.
5.

以上函数调用栈如下：

#0  index_build_storage 
#1  index_build
#2  index_create
#3  DefineIndex
#4  standard_ProcessUtility 
#5  gsaudit_ProcessUtility_hook
#6  pgaudit_ProcessUtility 
#7  hypo_utility_hook 
#8  ProcessUtility 
#9  PortalRunUtility
#10  PortalRun 
#11  exec_simple_query 
#12  PostgresMain1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.

分区表索引

创建普通表索引语法如下：

CREATE [ UNIQUE ] INDEX [ CONCURRENTLY ] [ [schema_name.]index_name ]
    { ON table_name [ USING method ] | [ USING method ] ON table_name }
    ({ { column_name | ( expression ) } [ COLLATE collation ] [ opclass ] [ ASC | DESC ] [ NULLS { FIRST | LAST } ] }[, ...] )
    [ index_option ]
    [ WHERE predicate ];1.
2.
3.
4.
5.

创建分区表索引语法：

CREATE [ UNIQUE ] INDEX [ [schema_name.]index_name ]
    { ON table_name [ USING method ] | [ USING method ] ON table_name }
    ( {{ column_name | ( expression ) } [ COLLATE collation ] [ opclass ] [ ASC | DESC ] [ NULLS LAST ] }[, ...] )
    [ LOCAL [ ( { PARTITION index_partition_name [ TABLESPACE index_partition_tablespace ] } [, ...] ) ] | GLOBAL ]
    [ index_option ]1.
2.
3.
4.
5.

两者执行流程基本一致，分区表索引在DefineIndex中是遍历每个分区调用partition_index_create。

文章转载自公众号：openGauss

分类

数据库

标签

openGauss

已于2023-5-5 11:00:33修改

51CTO

51CTO博客

51CTO学堂