ITPub博客

By my understanding the main difference between Sync and Async for VVR is how VVR writes VVR log and ship it to DR side. [@more@]Async mode, 1> Oracle messages OS to request IO operations to write oracle redo buffer into a certain group of log files when a client process commits a transaction.2> Oracle writes redo buffer into a certain group of log files. 3> Oracle completes the transaction and won’t wait for VVR. 4> VVR get the message from OS immediately and writes

http://www.dbspecialists.com/files/presentations/bulk_binds.html[@more@]

忙过了上一周上线的日子 终于可以喘口气了 总结一下目前发现的应用问题.1 还是发现了低效pl/sql2 还要加强对系统设计人员对于函数索引的培训...Load Profile~~~~~~~~~~~~ Per Second Per Transaction--------------- ---------------Redo size: 770,883.85 18,746.07Logical reads: 19,073.28 463.82Block changes: 3,656.68 88.92Physical reads: 698.60 16.99Physical writes: 278.54 6.77User calls: 1,313.17 31.93Parses: 703.90 17.12Hard parses: 18.46 0.45Sorts: 48.68 1.18Logons: 0.29 0.01Executes: 1,469.75 35.74Transactio

Is a long time since I wrote sth on Oracle last time,for I 've addressed myself to learning English from April this year.Now I come back and feel cool that English is not an obstacle for me to learn Oracle deeply ... ...Let's get to the point .[@more@]Oracle 10g后 默认将session_cached_cursors设置为20(之前为0) 表明了什么？Oracle建议在session中cache cursors.当然前提是应用中是sharable SQL(充分绑定变量)Why?大量重复执行相同sql可以:1 节省解析时间 2 降低cpu资源 If SESSION_CACHED_CURSORS is not set, it

bulk collect的效率极高，原因是批量操作大量减少了pl/sql引擎与sql引擎context的切换，降低cpu使用.而且通过合理设置limit,可以使得pga使用和undo空间使用非常节制Tom说过单条数据处理不能完成，不能满足需求的时候采用bulk collect是不错的选择.下面的例子处理600多万条记录的插入操作,应用此技术在1分半钟左右搞定.Oracle 9.2.0.6 64bit For AIX4C 6G存储一般ps.这次做了一个450G的数据迁移，用的就是RMAN duplicate DB ，而 bulk collect 做schema的拆分 11小时搞定schema的拆分[@more@]存储过程CREATE OR REPLACE PROCEDURE p_migrat

原贴发表在ITPUBhttp://www.itpub.net/581612.html试验环境:Linux AS4 + 10.2.0.1.0 - 64bit Production为了模拟数据库对象占用多个block,故设置较高pctfree.SQL> create table irs pctfree 80 tablespace users as 2 select rownum id,a.* from all_objects a;Table created.SQL> alter table irs add constraint irs_id_pk primary key(id); Table altered.SQL> create index irs_owner_idx on irs(owner) tablespace indx pctfree 80;Index created.SQL> show parameter optimizer_index NAME