[GE users] configure nodes to have a "head" node

davide cittaro daweonline at gmail.com
Mon Mar 20 16:29:43 GMT 2006

    [ The following text is in the "ISO-8859-1" character set. ]
    [ Your display is set for the "ISO-8859-10" character set.  ]
    [ Some special characters may be displayed incorrectly. ]

Hi Chris (sorry if I changed name/mail... I'm not morphing at all)

> Most people
> construct their cluster networks such that the compute nodes are not
> accessible by end users except through the head node.

This was the case up to last week (when we decided to give access to
nodes for special tasks...)

> Bio people
> tend to run lots of IO intensive serial jobs rather than slamming SGE
> with hundreds of thousands of small parallel tasks.

That is exactly what happens!

>  Your biggest
> performance bottleneck will be storage and fileservers if you are
> doing bioinformatics on this system.

That's right, we solved that and now we are serving 3 clusters without
any kind of filesystem bottleneck, actually it only happened that one
of our users started a batch script and generated a txt output file
that filled all the filesystem... Then we introduced quotas :)

I'm quite confident now that the right way would be setting one submit
node in addition to the others (option #1), I really appreciated your


To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
For additional commands, e-mail: users-help at gridengine.sunsource.net

More information about the gridengine-users mailing list