Re: [jira] [Created] (IOTDB-273) Parallel Recovery

[zyx]

Hi,


I would like to work on this issue.
My solusion is to parallelly recover the StorageGroupProcessor in 
StorageEngine. The degree of parallelism is the nmber of CPU cores and those 
thread could be managed by a fixed size thread pool.


---Original---
From: "Jialin Qiao (Jira)"<j...@apache.org&gt;
Date: Mon, Oct 28, 2019 11:15 AM
To: "dev"<dev@iotdb.apache.org&gt;;
Subject: [jira] [Created] (IOTDB-273) Parallel Recovery


Jialin Qiao created IOTDB-273:
---------------------------------

&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 
Summary: Parallel Recovery
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
 Key: IOTDB-273
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
 URL: https://issues.apache.org/jira/browse/IOTDB-273
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 
Project: Apache IoTDB
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Issue Type: Improvement
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Reporter: 
Jialin Qiao


When restarting IoTDB, we recover each storage group one by one, which costs 
20min for 8GB WAL.&nbsp;

The restart can speed up by parallelly recovery storage groups.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)