postgres源码解析54 Brin Index--1

Brin Index简介

brin index是Block range Index的缩写，顾名思义该索引是指块范围索引，该索引适合在超大表中进行过滤性扫描。基本的思路是追踪heap页域的最大值与最小值，用于过滤不符合条件的数据块。以下图为例，右边的堆表包含三个block，第一个block中有4个tuple，字段上的值分别是1、3、5、7。因此，与这个block相对应的Brin的元组就记录了 block的最小值1，最大值7。同理，2、4、6、8的最小值是2，最大值是8；12、11、14、13的最小值是10，最大值是14。这样就获得了一个 Brin index。

在执行带有条件的查询时，就可以利用Brin记录的这些信息，找到相应的block，同时滤掉大部分block，从而起到Index的作用。Brin Index是一种非常直观简单的Index。

关键数据结构

brin索引创建对应的handler，定义索引处理函数和属性

/*
 * BRIN handler function: return IndexAmRoutine with access method parameters
 * and callbacks.
 */
Datum
brinhandler(PG_FUNCTION_ARGS)
{
	IndexAmRoutine *amroutine = makeNode(IndexAmRoutine);

	amroutine->amstrategies = 0;
	amroutine->amsupport = BRIN_LAST_OPTIONAL_PROCNUM;
	amroutine->amcanorder = false;
	amroutine->amcanorderbyop = false;
	amroutine->amcanbackward = false;
	amroutine->amcanunique = false;
	amroutine->amcanmulticol = true;
	amroutine->amoptionalkey = true;
	amroutine->amsearcharray = false;
	amroutine->amsearchnulls = true;
	amroutine->amstorage = true;
	amroutine->amclusterable = false;
	amroutine->ampredlocks = false;
	amroutine->amcanparallel = false;
	amroutine->amcaninclude = false;
	amroutine->amkeytype = InvalidOid;

	amroutine->ambuild = brinbuild;
	amroutine->ambuildempty = brinbuildempty;
	amroutine->aminsert = brininsert;
	amroutine->ambulkdelete = brinbulkdelete;
	amroutine->amvacuumcleanup = brinvacuumcleanup;
	amroutine->amcanreturn = NULL;
	amroutine->amcostestimate = brincostestimate;
	amroutine->amoptions = brinoptions;
	amroutine->amproperty = NULL;
	amroutine->ambuildphasename = NULL;
	amroutine->amvalidate = brinvalidate;
	amroutine->ambeginscan = brinbeginscan;
	amroutine->amrescan = brinrescan;
	amroutine->amgettuple = NULL;
	amroutine->amgetbitmap = bringetbitmap;
	amroutine->amendscan = brinendscan;
	amroutine->ammarkpos = NULL;
	amroutine->amrestrpos = NULL;
	amroutine->amestimateparallelscan = NULL;
	amroutine->aminitparallelscan = NULL;
	amroutine->amparallelrescan = NULL;

	PG_RETURN_POINTER(amroutine);
}

Brin Index页格式

brin文件页有如下三种类型：元数据页，映射页与常规页类型
/* special space on all BRIN pages stores a "type" identifier */
#define		BRIN_PAGETYPE_META			0xF091
#define		BRIN_PAGETYPE_REVMAP		0xF092
#define		BRIN_PAGETYPE_REGULAR		0xF093

每个brin文件页含有一段特殊的空间BrinSpecialSpace，用于记录类型与标识信息
/*
 * Special area of BRIN pages.
 *
 * We define it in this odd way so that it always occupies the last
 * MAXALIGN-sized element of each page.
 */
typedef struct BrinSpecialSpace
{
	uint16		vector[MAXALIGN(1) / sizeof(uint16)];
} BrinSpecialSpace;

/* flags for BrinSpecialSpace */
#define		BRIN_EVACUATE_PAGE			(1 << 0)

通过如下宏定义从BrinSpecialSpace空间中获取页类型与标识信息
#define BrinPageType(page)		\
	(((BrinSpecialSpace *)		\
	  PageGetSpecialPointer(page))->vector[MAXALIGN(1) / sizeof(uint16) - 1])

#define BrinPageFlags(page)		\
	(((BrinSpecialSpace *)		\
	  PageGetSpecialPointer(page))->vector[MAXALIGN(1) / sizeof(uint16) - 2])

brin页类型的判断可通过调用如下宏实现：
#define BRIN_IS_META_PAGE(page) (BrinPageType(page) == BRIN_PAGETYPE_META)
#define BRIN_IS_REVMAP_PAGE(page) (BrinPageType(page) == BRIN_PAGETYPE_REVMAP)
#define BRIN_IS_REGULAR_PAGE(page) (BrinPageType(page) == BRIN_PAGETYPE_REGULAR)

void
brin_page_init(Page page, uint16 type)
{
	PageInit(page, BLCKSZ, sizeof(BrinSpecialSpace));
	BrinPageType(page) = type;
}

------------------------------------------------------------------------------------------------------------
void
PageInit(Page page, Size pageSize, Size specialSize)
{
	PageHeader	p = (PageHeader) page;

	specialSize = MAXALIGN(specialSize);

	Assert(pageSize == BLCKSZ);
	Assert(pageSize > specialSize + SizeOfPageHeaderData);

	/* Make sure all fields of page are zero, as well as unused space */
	MemSet(p, 0, pageSize);

	p->pd_flags = 0;
	p->pd_lower = SizeOfPageHeaderData;
	p->pd_upper = pageSize - specialSize;
	p->pd_special = pageSize - specialSize;
	PageSetPageSizeAndVersion(page, pageSize, PG_PAGE_LAYOUT_VERSION);
	/* p->pd_prune_xid = InvalidTransactionId;		done by above MemSet */
}