[kepler-users] globus toolkit 4: globus-ws submit actor.

Matthew Jones jones at nceas.ucsb.edu
Wed Apr 11 08:56:17 PDT 2007


Hi Terry,

There has been a lot of Grid support work in Kepler by various people. 
It has not been particularly well integrated (although see below).  Some 
people have put in support for NIMROD, others have worked on ssh-based 
job execution, and others have put in Globus actors.  The globus actors 
were written quite a while ago by Steve Mock at UCSD, and they target 
the Globus 2.x releases.  They include actors such as GlobusProxy, 
GlobusJob, and GridFTP.  I think these have not been tested in quite a 
while, and so I suspect it would take some tweaking to get them to work. 
  Also, I do not think anyone has put in GT4 support yet.  Some people 
at Monash University put in support for the Griddles toolkit, which 
allows submission and monitoring of globus jobs, but I'm not sure what 
version of globus.  Ilkay Altintas can probably provide more details 
about this work.

In addition, on the SEEK project we have Lucas Gilbert working on a new 
system for distributed execution of workflows, with the goal of 
providing a unifying framework for how distributed jobs are handled. 
Lucas can probably provide an update.

Hope this helps.

Matt

Terry.Rankine at csiro.au wrote:
> Hi
> 
> I am sure I asked this a long time ago but I cant get google to find it
> in your list archives.
> 
> I am interested in using/developing globus 4 toolkit actors for Kepler.
> 1. submittion. (gram ws)
> 2. information discovery (MDS4).
> 3. service monitoring.
> 4. rft service interface.
> 
> Do these actors exist? Are other groups interested? Is there funding
> allocated to useful actors?
> 
> Terry Rankine


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Matthew B. Jones
jones at nceas.ucsb.edu                Ph: 541-888-2581 ext. 287
National Center for Ecological Analysis and Synthesis (NCEAS)
UC Santa Barbara
http://www.nceas.ucsb.edu/ecoinformatics
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



More information about the Kepler-users mailing list