openGauss 内核分析（十）：数据库索引的创建过程

2023-07-27
北京
本文字数：3658 字
阅读完需：约 12 分钟

openGauss 内核分析（十）：数据库索引的创建过程

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            }        }

复制代码

#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"

复制代码

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。

zhang    lockmode = concurrent ? ShareUpdateExclusiveLock : ShareLock;    rel = heap_open(relationId, lockmode);

复制代码

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);

复制代码

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);

复制代码

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,……

复制代码

关闭表并返回索引表 id：

    heap_close(rel, NoLock);    return indexRelationId;

复制代码

Index_create函数

打开系统表 pg_class：

    pg_class = heap_open(RelationRelationId, RowExclusiveLock);

复制代码

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);

复制代码

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);

复制代码

将索引表元信息存入系统表 pg_index：

     UpdateIndexRelation(indexRelationId,        heapRelationId,        indexInfo,……

复制代码

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));    }

复制代码

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);

复制代码

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;}

复制代码

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:2834

复制代码

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  PostgresMain

复制代码

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 ];

复制代码

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 ]

复制代码

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。

发布于: 刚刚阅读数: 6

daydayup

关注

还未添加个人签名 2023-07-18 加入

还未添加个人简介

发布

暂无评论

创作场景

openGauss 内核分析（十）：数据库索引的创建过程

openGauss 内核分析（十）：数据库索引的创建过程

DefineInde

Index_create****函数

index_build

daydayup

评论

Index_create函数