[Bioclusters] High Availability Clustering

Ron Chen bioclusters@bioinformatics.org
Thu, 7 Aug 2003 21:10:09 -0700 (PDT)


Thanks!

Even I don't use PBSPro, I am happy that Altair is
still improving it. I am sure lots of PBS sites will
be excited about this feature.

Also, it would be great that you guys can backport
some of the fixes/enchancements to OpenPBS, while
adding new features like that into PBSPro.

 -Ron

--- Michael Humphrey <Humphrey@altair.com> wrote:
> Lai Loong Fong,
> 
> I assume you are using the term "shadow master" to
> refer to the ability to
> have the PBS Pro server and scheduler modules
> automatically migrate to a new
> hardware platform when the original master head node
> experiences a failure.
> 
> Currently PBS Pro server and scheduler modules can
> fail over when set-up on
> failsafe hardware. This current implementation
> requires special hardware and
> operating system support from hardware
> manufacturers.
> 
> This new software failover capability is currently
> in beta testing within
> PBS Pro and we are planning to make this new
> functionality available in
> Version 5.4 of PBS Pro to be announced in November
> of this year. This new
> functionality will not require special hardware
> environments.
> 
> Mike
> 
> PS: NCSA is currently considering migrating from
> OpenPBS to PBS Pro. 
> 
> -----Original Message-----
> From: LAI Loong Fong
> [mailto:lailf@bii.a-star.edu.sg] 
> Sent: Thursday, July 24, 2003 9:29 PM
> To: Bioclusters; Robert Walsh; Michael Humphrey
> Subject: Re: [Bioclusters] High Availability
> Clustering
> 
> 
> I am not sure whether NCSA or NCAR are using
> Scalable PBS, the only info
> that I got from them is that they are not using the
> standard OpenPBS &
> PBSPro. I agree with you the problem with PBSes not
> having "shadow master",
> anyway Michael will be in Singapore next week and I
> will be meeting up with
> him, so I will raise this point with him. Any other
> things that you guys
> like me to ask Michael about?
> 
> Cheers
> LAI Loong Fong
> 
> On 25/7/03 9:07 AM, "Ron Chen"
> <ron_chen_123@yahoo.com> wrote:
> 
> > Are you talking about Scalable PBS?
> > 
> > One important feature that is still missing is the
> > "shadow master", it is simple to implement, but
> AFAIK,
> > PBS, OpenPBS, and ScalablePBS do not have it.
> > 
> > NOT that I wanted to suggest GridEngine everytime,
> but
> > if PBS is still missing this feature, I don't
> think I
> > would use PBS in any production cluster!
> > 
> > (for those on the PBS mailing list, you would know
> > what I am talking about)
> > 
> > -Ron
> > 
> > --- LAI Loong Fong <lailf@bii.a-star.edu.sg>
> wrote:
> >> That's because NCSA and also NCAR are using a
> hacked
> >> version of PBS rather
> >> than the one that you can download.
> >> 
> >> LAI Loong Fong
> >> 
> >> _______________________________________________
> >> Bioclusters maillist  -
> >> Bioclusters@bioinformatics.org
> >> 
> >
>
https://bioinformatics.org/mailman/listinfo/bioclusters
> > 
> > 
> > __________________________________
> > Do you Yahoo!?
> > Yahoo! SiteBuilder - Free, easy-to-use web site
> design software 
> > http://sitebuilder.yahoo.com 
> > _______________________________________________
> > Bioclusters maillist  - 
> Bioclusters@bioinformatics.org 
> >
>
https://bioinformatics.org/mailman/listinfo/bioclusters
> > 
> _______________________________________________
> Bioclusters maillist  - 
> Bioclusters@bioinformatics.org
>
https://bioinformatics.org/mailman/listinfo/bioclusters


__________________________________
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software
http://sitebuilder.yahoo.com