Thank you very much.

Fei

> On Jan 14, 2015, at 7:36 PM, Ted Yu <[email protected]> wrote:
> 
> https://issues.apache.org/jira/browse/HDFS-7613 
> <https://issues.apache.org/jira/browse/HDFS-7613>
> 
> Cheers
> 
> On Wed, Jan 14, 2015 at 4:35 PM, Fei Hu <[email protected] 
> <mailto:[email protected]>> wrote:
> Hi,
> 
> Thank you for your help.
> 
> I searched HDFS-7613 by Google and the link 
> https://issues.apache.org/jira/issues/?jql=project%20%3D%20HDFS%20AND%20text%20~%20%227613%22
>  
> <https://issues.apache.org/jira/issues/?jql=project%20=%20HDFS%20AND%20text%20~%20%227613%22>,
>  but I could not find it.
> 
> Could you email me the link? Thank you very much.
> 
> Sincerely,
> Fei 
> 
> 
>> On Jan 14, 2015, at 5:43 PM, Ted Yu <[email protected] 
>> <mailto:[email protected]>> wrote:
>> 
>> Fei:
>> You can watch this issue:
>> HDFS-7613 Block placement policy for erasure coding groups
>> 
>> the solution there would be helpful to us.
>> 
>> Cheers
>> 
>> On Wed, Jan 14, 2015 at 11:04 AM, Fei Hu <[email protected] 
>> <mailto:[email protected]>> wrote:
>> Thank you for your quick response.
>> 
>> After reading the materials you recommended, my conclusion is that Hadoop 
>> does not provide interface to customize the data placement policy. We need 
>> to add some codes to the source package of HDFS. Is that right?
>> 
>> Thanks,
>> Fei
>> 
>> On Tue Jan 13 2015 at 10:42:27 PM Ted Yu <[email protected] 
>> <mailto:[email protected]>> wrote:
>> See this thread: http://search-hadoop.com/m/lAh9i28K7 
>> <http://search-hadoop.com/m/lAh9i28K7>
>> 
>> See also HDFS-7228
>> 
>> Cheers
>> 
>> On Tue, Jan 13, 2015 at 7:33 PM, Fei Hu <[email protected] 
>> <mailto:[email protected]>> wrote:
>> Hi,
>> 
>> I want to customize the data placement strategy rather than using the 
>> default strategy in HDFS. Is there any way to control which datanode the 
>> replica is delivered to?
>> 
>> Thank you in advance.
>> 
>> Best regards,
>> Fei
>> 
>> 
> 
> 

Reply via email to