On Tue, Jun 06, 2006 at 02:08:50PM +0100 or thereabouts, Jensen, J (Jens) wrote:
> Yes, eventually dCache at Tier 1 will be phased out.
>
> Nevertheless, we will continue to support dCache as long as
> Tier 2s are running dCache - subject to funding from GridPP
> and having enough people - at least we intend to support it
> as much as we can.
>
> In fact we can use CASTOR today; at least the SRM appears
> to be working and our friends in CMS will be brea^W testing
> it Really Soon Now(tm). Data is not kept, though, we're
> still testing.
>
> Anyone wants to have a go, let me know. There's a dteam
> allocation too.
>
> Now I am curious why "effectively" srmcp can't be used with
> CASTOR? It works fine for me. Anyone knows or should I
> ask Lassi?
Looked at this last point yesterday and today. castor-srmv1 does
of course not support SRMCP(*) so the srmcp tool can only be used
where one of your transfer supports push or pull mode.
Consequently it is impossible to use srmcp between say CERN castor
and RAL castor.
* SRMCP = SRM protocol method.
srmcp = d-cache provide transfer client.
Steve
>
> Cheers,
> --jens
>
>
> PS. Quite enjoyed "WLCG is not easily moved by technical facts"
> :-)
>
> -----Original Message-----
> From: GRIDPP2: Deployment and support of SRM and local storage
> management [mailto:[log in to unmask]]On Behalf Of Greig A
> Cowan
> Sent: 05 June 2006 10:49
> To: [log in to unmask]
> Subject: Re: FW: FTS and Phedex
>
>
> > " With RAL moving to Castor " RAL aren't getting rid of their dcache are they?
>
> Yep, they sure are.
>
> They should be using CASTOR by the end of the year I think. Jens can tell
> you more.
>
> Greig
>
> >
> > On 05/06/06, Olivier van der Aa <[log in to unmask]> wrote:
> > > FYI, Lassi could not cc to the storage list.
> > >
> > > Cheers, Olivier.
> > >
> > > -------- Original Message --------
> > > Subject: Re: Disabled Castor/Castor load test transfers
> > > Date: Sat, 3 Jun 2006 17:31:14 +0200
> > > From: Lassi A. Tuura <[log in to unmask]>
> > > To: [log in to unmask]
> > >
> > >
> > > *** Discussion title: PhEDEx Discussions
> > >
> > > Hi,
> > >
> > > > The iowait we observed happened when we where doing dCache->dCache
> > > > transfers using fts.
> > > >
> > > > You can see the iowait difference between srmcp and fts on this page
> > > > http://www.gridpp.ac.uk/wiki/IC-HEP#Transfer_Tests_2006-02-22
> > > >
> > > > We compared how the data is writen on the disk when using fts or srmcp
> > > >
> > > > http://www.hep.ph.ic.ac.uk/~georgiou/dcache_fts/fts-0.png
> > > > http://www.hep.ph.ic.ac.uk/~georgiou/dcache_fts/srm-0.png
> > > >
> > > > you can see the difference yourself. Basically using fts causes the
> > > > disk to jump when writing while srm doing fine.
> > >
> > > Thanks Olivier. Tim Barrass circulated these notes around before, so
> > > we are aware of the situation.
> > >
> > > > The transfer rate we could achieve using srmcp (with phedex) was
> > > > close to 400Mb/s while with fts we never reached more than 140Mb/s
> > > > for dcache->dcache transfers.
> > > >
> > > > More tests haven been achieved by configuring the fts channel to
> > > > use srmcp rather than gsiftp third party copy. It improved the
> > > > situation but not dramatically. There will be an upgrade to 1.5 at
> > > > ral next monday and I hope we can make another set of tests.
> > >
> > > I understand. We still need to move to using FTS. The weekly WLCG
> > > service meetings are a good place to raise your concerns about the
> > > performance. You should make sure Ian / Micheal / Stefano raise that
> > > issue as they participate in those meetings, but you should be aware
> > > that WLCG is not easily moved by technical facts.
> > >
> > > Please do keep in mind that we effectively cannot use srmcp against
> > > Castor/SRM at this time, and most sites do not respond to problem
> > > reports concerning srmcp, or do nothing about them. The reality is
> > > that in practice WLCG only supports the subset of SRM used by FTS.
> > > With RAL moving to Castor you would do well exercise your technical
> > > options earlier rather than later so you can back your position with
> > > clearly articulated numbers.
> > >
> > > Lassi
> > >
> > > -------------------------------------------------------------
> > > Visit this CMS message (to reply or unsubscribe) at:
> > > https://hypernews.cern.ch/HyperNews/CMS/get/phedex/207/4/1/1/1/1/1/2/1.html
> > >
> > > --
> > > - O. van der Aa - Imperial College London -
> > > - LT2 Technical Coordinator -
> > > - tel: +442075947810, +442071005426 -
> > > - SIP: [log in to unmask] -
> > > - fax: +442078238830 -
> > > - http://surl.se/agtu -
> > >
> >
>
> --
> ========================================================================
> Dr Greig A Cowan http://www.ph.ed.ac.uk/~gcowan1
> School of Physics, University of Edinburgh, James Clerk Maxwell Building
>
> TIER-2 STORAGE SUPPORT PAGES: http://wiki.gridpp.ac.uk/wiki/Grid_Storage
> ========================================================================
--
Steve Traylen
[log in to unmask]
http://www.gridpp.ac.uk/
|