增加jobhitory缓存详解 (#27)
Reviewed-on: #27 Co-authored-by: zeekling <lingzhaohui@zeekling.cn> Co-committed-by: zeekling <lingzhaohui@zeekling.cn>
This commit is contained in:
parent
f3917672d1
commit
3d1a30e0de
75
yarn/jobhistory_cache.md
Normal file
75
yarn/jobhistory_cache.md
Normal file
@ -0,0 +1,75 @@
|
|||||||
|
|
||||||
|
## jobhistory 作业缓存
|
||||||
|
|
||||||
|
jobhistory 一般会保存一部分作业信息到内存中,查询作业信息的时候一般会从内存查询,如果内存查询不到就会从磁盘上扫描。
|
||||||
|
|
||||||
|
jobhistory 缓存一般分为两层,第一层是guava缓存,默认情况下guava的缓存个数是5,可以通过配置项`mapreduce.jobhistory.loadedjobs.cache.size`控制。
|
||||||
|
|
||||||
|
当guava的一级缓存中不存在的时候,默认是需要重新加载的,jobhistory中定义了加载规则,定义代码如下:
|
||||||
|
|
||||||
|
```java
|
||||||
|
CacheLoader<JobId, Job> loader;
|
||||||
|
loader = new CacheLoader<JobId, Job>() {
|
||||||
|
@Override
|
||||||
|
public Job load(JobId key) throws Exception {
|
||||||
|
return loadJob(key);
|
||||||
|
}
|
||||||
|
};
|
||||||
|
```
|
||||||
|
|
||||||
|
其中loadJob实现如下,其中hsManager为加载具体实现,
|
||||||
|
|
||||||
|
```java
|
||||||
|
private Job loadJob(JobId jobId) throws RuntimeException, IOException {
|
||||||
|
if (LOG.isDebugEnabled()) {
|
||||||
|
LOG.debug("Looking for Job " + jobId);
|
||||||
|
}
|
||||||
|
HistoryFileInfo fileInfo;
|
||||||
|
|
||||||
|
fileInfo = hsManager.getFileInfo(jobId);
|
||||||
|
|
||||||
|
if (fileInfo == null) {
|
||||||
|
throw new HSFileRuntimeException("Unable to find job " + jobId);
|
||||||
|
}
|
||||||
|
|
||||||
|
fileInfo.waitUntilMoved();
|
||||||
|
|
||||||
|
if (fileInfo.isDeleted()) {
|
||||||
|
throw new HSFileRuntimeException("Cannot load deleted job " + jobId);
|
||||||
|
} else {
|
||||||
|
return fileInfo.loadJob();
|
||||||
|
}
|
||||||
|
}
|
||||||
|
```
|
||||||
|
|
||||||
|
hsManager中定义了jobhistory的二级缓存:jobListCache,jobListCache的大小可以通过配置项`mapreduce.jobhistory.joblist.cache.size`控制。
|
||||||
|
默认可以保存20000个。当然缓存超时指定时间可会被清理,具体可以有配置项`mapreduce.jobhistory.max-age-ms`控制,默认为1周。
|
||||||
|
|
||||||
|
查找的顺序为:
|
||||||
|
- 优先从内存查找(二级缓存),为jobListCache。
|
||||||
|
- 如果缓存找不见,优先扫描刚完成的作业所在的目录,会刷新jobListCache缓存,由配置项mapreduce.jobhistory.intermediate-done-dir控制。
|
||||||
|
- 如果还是找不见,从已经完成的作业的目录扫描,具体目录由配置项mapreduce.jobhistory.done-dir控制。
|
||||||
|
|
||||||
|
```java
|
||||||
|
public HistoryFileInfo getFileInfo(JobId jobId) throws IOException {
|
||||||
|
// 优先从内存查找(二级缓存)
|
||||||
|
HistoryFileInfo fileInfo = jobListCache.get(jobId);
|
||||||
|
if (fileInfo != null) {
|
||||||
|
return fileInfo;
|
||||||
|
}
|
||||||
|
// 如果缓存找不见,优先扫描刚完成的作业所在的目录,由配置项mapreduce.jobhistory.intermediate-done-dir控制
|
||||||
|
scanIntermediateDirectory();
|
||||||
|
fileInfo = jobListCache.get(jobId);
|
||||||
|
if (fileInfo != null) {
|
||||||
|
return fileInfo;
|
||||||
|
}
|
||||||
|
|
||||||
|
// 如果还是找不见,从已经完成的作业的目录扫描,具体目录由配置项mapreduce.jobhistory.done-dir控制
|
||||||
|
fileInfo = scanOldDirsForJob(jobId);
|
||||||
|
if (fileInfo != null) {
|
||||||
|
return fileInfo;
|
||||||
|
}
|
||||||
|
return null;
|
||||||
|
}
|
||||||
|
```
|
||||||
|
|
Loading…
Reference in New Issue
Block a user