【转载】latch: cache buffers chains

  

    当一个数据块读入sga区，相应的buffer header会被放置到hash列表上,我们称其这hash chains,chain在中文的意为链条或串的意思，表达就是关连性.如果一个进程想访问或修改hash chain上的block,它首先要获得”cache buffers chains” latch。

 

原因一：低效率的SQL语句（主要体现在逻辑读过高）

cache buffers chains latch很大程度与逻辑读有关，所以要观注v$sql中BUFFER_GETS/EXECUTIONS大的语句。
同时每一个逻辑读需要一个latch get 操作及一个cpu操作，这样的sql也会很耗cpu资源。

 

原因二：热块（访问过于频繁）

找出热点块方法一：

--找出p1raw

select p1,p1raw from v$session_wait where event='latch: cache buffers chains';

--找到对象

SELECT /*+ RULE */

 E.OWNER || '.' || E.SEGMENT_NAME SEGMENT_NAME,

 E.PARTITION_NAME,

 E.EXTENT_ID EXTENT#,

 X.DBABLK - E.BLOCK_ID + 1 BLOCK#,

 X.TCH,

 L.CHILD#

  FROM SYS.V$LATCH_CHILDREN L, SYS.X$BH X, SYS.DBA_EXTENTS E

 WHERE X.HLADDR = '00000002576EE018'--p1raw

   AND E.FILE_ID = X.FILE#

   AND X.HLADDR = L.ADDR

   AND X.DBABLK BETWEEN E.BLOCK_ID AND E.BLOCK_ID + E.BLOCKS - 1

 ORDER BY X.TCH DESC;

找出热点块方法二：

--直接找出热点块

SELECT OBJECT_NAME, SUBOBJECT_NAME

  FROM DBA_OBJECTS

 WHERE DATA_OBJECT_ID IN

       (SELECT DATA_OBJECT_ID

          FROM (SELECT OBJ DATA_OBJECT_ID, FILE#, DBABLK, CLASS, STATE, TCH

                  FROM X$BH

                 WHERE HLADDR IN (SELECT ADDR

                                    FROM (SELECT ADDR

                                            FROM V$LATCH_CHILDREN

                                           ORDER BY (GETS + MISSES + SLEEPS) DESC)

                                   WHERE ROWNUM < 10)

                 ORDER BY TCH DESC)

         WHERE ROWNUM < 10);