hadoop-common-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Juwei Shi <shiju...@gmail.com>
Subject Re: A problem about the network
Date Mon, 18 Apr 2011 13:44:00 GMT
Did you look at the cluster wide resource usage such as CPU, memory and
network I/O? You may check the resouce utilization when the processing
becomes slow. Then the bottleneck can be detected.

在 2011年4月18日 下午9:38,ajing.wang <ajing.wang@qq.com>写道:

> HI,all.
>   I have encountered  a problem about hardware .  I build up a hadoop
> cluster with  four old computer. After start the cluster and running job for
>  few days , I found the computer's network become slow,and the job's process
> time become longer and longer. It becomes normal after rebooting the
> computer.
>   I asked the system manager,they said  the  network card may be not so
> good.
>   have you encounter the problem like this? and How to solve this question?
>   Thanks.

- Juwei

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message