airavata-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Pamidighantam, Sudhakar V" <spami...@illinois.edu>
Subject Re: [jira] [Commented] (AIRAVATA-1635) [GSoC] Integrate Airavata Java Client SDK with GridChem Client
Date Sat, 28 Mar 2015 13:42:55 GMT
In either case the file is not changed on the client side so that is not of any significance.
The migration to Airavata is in reducing the maintenance of requirement defining code in multiple
locations ( such as Airavata Server and HPC Host etc…) into once location while managing
all the relevant requirements and validation. Some of the validation can/should be done at
the client itself and this also needs strengthening.

 But your project will focus on three modular steps each of which could be independent but
influence the others.
1. authentication migration to Airavata Identity service 
2. migration of application execution to Airavata Job/Workflow management 
3. migration of ticketing and notification system to Jira. 

Ticketing system requires authentication so they are coupled. Authentication is coupled to
user registration. Of course job executions require authentication as well. So this should
be the first one to look at.  

So we should look at the issues carefully and think how a production system can be migrated
without/with only minimal disruption. 

Sudhakar.
On Mar 27, 2015, at 11:05 PM, DImuthu Upeksha <dimuthu.upeksha2@gmail.com> wrote:

> Hi Sudhakar,
> 
> So if we port this use case to Airavata, we don't have to change the file at client side.
Only requirement is to pass this file to Airavata through its API. Am I correct?
> 
> On Fri, Mar 27, 2015 at 5:23 PM, Pamidighantam, Sudhakar V <spamidig@illinois.edu>
wrote:
> Dimuthu:
> This is a standard format for an application named Gaussian. This file is parsed for
various job requirements at the middleware and resource stages for proper job submission.
There are several applications each of which have their own standard ways of setting the job
parameters. Also these files are used by the applications for setting variables internally
when they execute.
> 
> 
> Sudhakar.
> 
> 
> On Mar 27, 2015, at 4:05 AM, Dimuthu Upeksha (JIRA) <jira@apache.org> wrote:
> 
> >
> >    [ https://issues.apache.org/jira/browse/AIRAVATA-1635?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14383553#comment-14383553
]
> >
> > Dimuthu Upeksha commented on AIRAVATA-1635:
> > -------------------------------------------
> >
> > I tried GridChem client by putting some jobs on the servers and getting outputs.
Input files that we push has a format like this.
> >
> > %chk=water.chk
> > %nprocshared=1
> > %mem=500MB
> > #P RHF/6-31g* opt pop=reg gfinput gfprint iop(6/7=3) SCF=direct
> >
> > Gaussian Test Job 00
> > Water with archiving
> >
> > 0 1
> > O
> > H 1 0.96
> > H 1 0.96 2 109.471221
> >
> > Is this format some standard way of passing jobs or specific format for GridChem?
> > Does client directly pass this file to middleware or parse this and pass only necessary
data?
> >
> >> [GSoC] Integrate Airavata Java Client SDK with GridChem Client
> >> ---------------------------------------------------------------
> >>
> >>                Key: AIRAVATA-1635
> >>                URL: https://issues.apache.org/jira/browse/AIRAVATA-1635
> >>            Project: Airavata
> >>         Issue Type: Epic
> >>           Reporter: Suresh Marru
> >>             Labels: gsoc, gsoc2015, mentor
> >>
> >> GridChem is a Science Gateway enables users to run computational experiments
on multiple supercomputing resources. Currently GridChem, a java swing based webstart client
[1] uses a Axis2 based Middleware Service [2] which brokers users actions into computational
jobs.
> >> This project needs to understand the Client [1] and port it to use Apache Airavata
java client SDK. The project has following components:
> >> * Integrate GridChem client with Airavata User Store (implemented by WSO2 Identity
Server)
> >> * Integrate with Airavata API for application executions.
> >> * Integrate with Atlassian JIRA + Confluence for user error reporting and status
notifications.
> >> [1] - https://github.com/SciGaP/GridChem-Client
> >> [2] - https://github.com/SciGaP/GridChem-Middleware-Service
> >
> >
> >
> > --
> > This message was sent by Atlassian JIRA
> > (v6.3.4#6332)
> >
> 
> 
> 
> 
> -- 
> Regards
> W.Dimuthu Upeksha
> Undergraduate
> Department of Computer Science And Engineering
> University of Moratuwa, Sri Lanka


Mime
View raw message