hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Xiaobing Zhou (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HDFS-10431) Refactor Async DFS related tests to reuse shared instance of AsyncDistributedFileSystem instance to speed up tests
Date Wed, 25 May 2016 23:25:12 GMT

     [ https://issues.apache.org/jira/browse/HDFS-10431?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Xiaobing Zhou updated HDFS-10431:
---------------------------------
    Description: 
1. Move irrelevant parts out of TestAsyncDFSRename
2. Limit of max async calls(i.e. ipc.client.async.calls.max) is set and cached in ipc.Client.
Client instances are cached based on SocketFactory. In order to test different cases in various
limits, every test (e.g. TestAsyncDFSRename and TestAsyncDFS) creates separate instance of
MiniDFSCluster and that of AsyncDistributedFileSystem hence. This is not efficient in that
tests may take long time to bootstrap MiniDFSClusters. It's even worse if cluster needs to
restart in the middle. This proposes to do refactoring to use shared instance of AsyncDistributedFileSystem
for speedup.

  was:Limit of max async calls(i.e. ipc.client.async.calls.max) is set and cached in ipc.Client.
Client instances are cached based on SocketFactory. In order to test different cases in various
limits, every test (e.g. TestAsyncDFSRename and TestAsyncDFS) creates separate instance of
MiniDFSCluster and that of AsyncDistributedFileSystem hence. This is not efficient in that
tests may take long time to bootstrap MiniDFSClusters. It's even worse if cluster needs to
restart in the middle. This proposes to do refactoring to use shared instance of AsyncDistributedFileSystem
for speedup.


> Refactor Async DFS related tests to reuse shared instance of AsyncDistributedFileSystem
instance to speed up tests
> ------------------------------------------------------------------------------------------------------------------
>
>                 Key: HDFS-10431
>                 URL: https://issues.apache.org/jira/browse/HDFS-10431
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>          Components: hdfs
>            Reporter: Xiaobing Zhou
>            Assignee: Xiaobing Zhou
>
> 1. Move irrelevant parts out of TestAsyncDFSRename
> 2. Limit of max async calls(i.e. ipc.client.async.calls.max) is set and cached in ipc.Client.
Client instances are cached based on SocketFactory. In order to test different cases in various
limits, every test (e.g. TestAsyncDFSRename and TestAsyncDFS) creates separate instance of
MiniDFSCluster and that of AsyncDistributedFileSystem hence. This is not efficient in that
tests may take long time to bootstrap MiniDFSClusters. It's even worse if cluster needs to
restart in the middle. This proposes to do refactoring to use shared instance of AsyncDistributedFileSystem
for speedup.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-help@hadoop.apache.org


Mime
View raw message