You are viewing a plain text version of this content. The canonical link for it is here.

Posted to common-user@hadoop.apache.org by qi wu <ch...@gmail.com> on 2007/07/04 11:57:00 UTC

Can running multiple map/reduce jobs in single machine improve the performance ?

Hi,
Currently,I am using nutch in a single Linux box with 1G memory and one 3GHZ Intel P4 CPU. The hadoop is in local mode with only one  map reduce job running concurrently. and the map reduce  process is very slow. I am wondering whether the solution below can improve the performance ?
1. Run the hadoop in  distributed mode,and use more than map/reduce jobs in one machine?
2. Increase the memory size ?
Any suggestion for improving the performance are welcome ! Thanks in advance!!  

-chee