Re: How to stop the nodes but leave the master running?

From: Robert Yu <no email>
Date: Wed, 19 Jun 2013 13:32:04 -0700

Hi,
Thanks, but that's not what I had in mind. I do not want to *terminate*
the idle nodes. I want to "stop" them, which is like putting them to
sleep. Waking them up again ("start -x") is faster than creating new
instances.
Thanks,
Robert


On Wed, Jun 19, 2013 at 1:25 PM, Rajat Banerjee <rajatb_at_post.harvard.edu>wrote:

> You can use the elastic load balancer to shut down the slave nodes as
> their jobs run out. Here's the document on it:
>
> http://star.mit.edu/cluster/docs/0.93.3/manual/load_balancer.html
>
>
> On Tue, Jun 18, 2013 at 10:45 PM, Robert Yu <robert.yu_at_aditazz.com> wrote:
>
>> Hi,
>>
>> I would like to stop the nodes so that I can save money and restart
>> (start -x) them later on. I am also using
>> sshfs to access the shared volume, so I would like to keep the master
>> alive. How do I do this from the command
>> line? I know how to do this from the AWS web console.
>>
>> Thanks,
>> Robert
>>
>> --
>> Robert Yu, Member Technical Staff
>>
>> www.aditazz.com | robert.yu_at_aditazz.com
>> 1111 Bayhill Drive Suite 260 | San Bruno | CA 94066
>> 510.459.0216 | cell
>> 650.627.7357 | 650.492.7000 x1008 | work
>> 650.684.1149 | fax
>>
>>
>> _______________________________________________
>> StarCluster mailing list
>> StarCluster_at_mit.edu
>> http://mailman.mit.edu/mailman/listinfo/starcluster
>>
>>
>

-- 
Robert Yu, Member Technical Staff
www.aditazz.com | robert.yu_at_aditazz.com
1111 Bayhill Drive Suite 260 | San Bruno | CA 94066
510.459.0216 | cell
650.627.7357 | 650.492.7000 x1008 | work
650.684.1149 | fax

Received on Wed Jun 19 2013 - 16:32:37 EDT