Home | About | Sematext search-lucene.com search-hadoop.com
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB
 Search Hadoop and all its subprojects:

Switch to Threaded View
HDFS >> mail # dev >> A question about the namenode's decision on the data placement of a new file


Copy link to this message
-
Re: A question about the namenode's decision on the data placement of a new file
On Fri, Aug 24, 2012 at 3:33 PM, wei xu <[EMAIL PROTECTED]> wrote:
> I am doing some research on the data placement, but I am not quite familiar
> with the Hadoop, is there any one who can tell me when a new file be added
> into the HDFS, which function will be called by namenode to make a decision
> on the allocation ( I mean, if there has a list of datanodes and the file
> might be larger than 64MB, that is, several blocks may be included. What's
> the function that decide the placement of those blocks to those datanode?)

You'll want to read BlockPlacementPolicy and its inheritors.

-andy
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB