Know more about checkpoint

最新推荐文章于 2024-10-16 11:14:53 发布

iteye_4537

最新推荐文章于 2024-10-16 11:14:53 发布

阅读量97

点赞数

文章标签：数据库数据结构与算法

Know more about checkpoint checkpoint 分成很多种 full 、file、thread、parallel query、 object 、incremental 、logfile switch

每一种checkpoint 都有其自身的特性，例如Incremental Checkpoint会要求ckpt 每3s 更新一次controlfile 但是不更新datafile header，而FULL CHECKPOINT要求立即完成(同步的) 且会同时更新 controlfile 和 datafile header。

各种checkpoint 的特点见下表：

Full Checkpoint Writes block images to the database for all dirty buffers from all instances Statistics updated: DBWR checkpoints DBWR checkpoint buffers written DBWR thread checkpoint buffers written Caused by: Alter system checkpoint [global] Alter database begin backup Alter database close Shutdown Controlfile and datafile headers are updated CHECKPOINT_CHANGE# Thread Checkpoint Writes block images to the database for all dirty buffers from one instance Statistics updated: DBWR checkpoints DBWR checkpoint buffers written DBWR thread checkpoint buffers written Caused by: Alter system checkpoint local Controlfile and datafile headers are updated CHECKPOINT_CHANGE# File Checkpoint Writes block images to the database for all dirty buffers for all files of a tablespace from all instances Statistics updated: DBWR tablespace checkpoint buffers written DBWR checkpoint buffers written DBWR checkpoints Caused by: Alter tablespace XXX offline Alter tablespace XXX begin backup Alter tablespace XXX read only Controlfile and datafile headers are updated CHECKPOINT_CHANGE# Parallel Query Checkpoint Writes block images to the database for all dirty buffers belonging to objects accessed by the query from all instances Statistics updated: DBWR checkpoint buffers written DBWR checkpoints Caused by: Parallel Query Parallel Query component of PDML or PDDL Mandatory for consistency Object “Checkpoint” Writes block images to the database for all dirty buffers belonging to an object from all instances Statistics updated: DBWR object drop buffers written DBWR checkpoints Caused by: Drop table XXX Drop table XXX purge Truncate table XXX Mandatory for media recovery purposes Incremental Checkpoint Writes the contents of “some” dirty buffers to the database from CKPT-Q Block images written in SCN order Checkpoint RBA updated in SGA Statistics updated: DBWR checkpoint buffers written Controlfile is updated every 3 seconds by CKPT Checkpoint progress record Log Switch Checkpoint(8i 以前 LOG switch checkpoint是FULL CHECKPOINT) Writes the contents of “some” dirty buffers to the database Statistics updated: DBWR checkpoints DBWR checkpoint buffers written background checkpoints started background checkpoints completed Controlfile and datafile headers are updated CHECKPOINT_CHANGE#

无论是什么类型的checkpoint 检查点，所有的本地检查点(CKPT)已类似的本地化方法处理。每一个实例中所有本地的活跃检查点请求(active local checkpoint request)都保存在一个队列(queue)中，这个队列叫做 Active Checkpoint Queue。在这个队列(queue)中的每一条记录代表一个本地检查点(local checkpoint request)。当某一个进程( 可能是前台进程 foreground process --例如前台进程执行“alter tablespace users begin/end backup"，也可能是CKPT 或者其他后台进程) ，这个进程都会将新的 request 记录放到这个Active Checkpoint Queue中。典型的一个checkpoint request 由检查点类型checkpoint request type，优先级priority , 以及与该checkpoint request 关联的checkpoint structure，等待进程 waiter process，还有其余一些相关的属性如 FILE Checkpoint 的tablespace id、FILE NUMBER、 Object checkpoint的object id 等。

DBWR进程会不断地扫描这个Active Checkpoint Queue，并服务于这个Queue上的checkpoint request 检查点请求。一旦某个request 被完成了，DBWR 将这个request标记为completed 。 CKPT 进程也会不断监控这个 Active Checkpoint Queue 查看是否所有request都被完成了。当CKPT发觉一个checkpoint request完成了， CKPT会将这个request从 Active Checkpoint Queue中移除。取决于不同的检查点种类和目的，当一个本地检查点(local checkpoint)完成，这意味着某些特定的磁盘上的数据结构被更新，以反映这个检查点完成的物理表现。这个操作或者由直接CKPT完成，或者由提交checkpoint request的等待进程直接完成，或者由CKPT唤醒这个提交checkpoint request的等待进程间接地完成，以上具体由谁来完成操作取决于检查点种类和目的。