Docstoc
EXCLUSIVE OFFER FOR DOCSTOC USERS
Try the all-new QuickBooks Online for FREE.  No credit card required.

talk

Document Sample
talk Powered By Docstoc
					State of Play:
The Annual DSG Review

 Mark Baker
 The Distributed Systems Group
 University of Portsmouth,
 http://dsg.port.ac.uk/mab/

                                 1
    Outline
   Introduction.
   DSG Activities:
       OGSA Testbed,
       MPJ,
       GridRM,
       JEM
       UISB,
       jGMA,
       Others.
   Group Performance.
   SWOT.
   Summary.

28 October, 2004        Mark.Baker@computer.org   2
    Introduction
   Aims of the talk are:
       Present the annual review of the DSG, what has
        been done and where we are going…
       Have a look at some of the strengths and
        weaknesses of the group.
       Cross reference what we said we would do and
        what we have done!
       Try to take a wider view of future research and
        where we will be five years downstream!




28 October, 2004     Mark.Baker@computer.org          3
The Distributed Systems Group


    The Last Year’s Activities




                                 4
    OGSA Testbed
   One year project – Sep 2003 – Oct 2004.
   With Daresbury, Manchester, Reading and
    Westminster.
   Aimed to install GTx at all sites and deploy
    apps across testbed.
   Hong (funded) and Mat worked on the
    effort.
   Regular testbed meetings – at Westminster.
   Fund with Starbug + firewalls…


28 October, 2004   Mark.Baker@computer.org     5
    OGSA Testbed - Outcomes
   Good Aspects:
       Knowledge and learning:
          Hong at GoSC event today,

          SC Demos (jGMA+ GridRM),

          Papers at AHM (3 + poster).

       Integration with the wider e-Science programme,
       Status and profile – e-Science, Globus Alliance…
       Useful platform for testing – over the learning curve!,
       Leverage for future collaborations and proposals,
       Even though funding over, testbed partners want to
        continue working together.
       Won Web site bronze prize at AHM!
          Oh and the case of wine…




28 October, 2004         Mark.Baker@computer.org                  6
    OGSA Testbed - Outcomes
   Not such good aspects:
       Painful process, working with less than stable
        software,
       Hiccup and change of direction in Jan 04,
       What did the “other” testbed actually do?
       No following on funding:
            GT4 effort within ETF – we only have a “watching”
             brief!
       Firewalls – remember GridFTP!
       Where to place our efforts with this, in the
        future.

28 October, 2004         Mark.Baker@computer.org            7
    MPJ: Progress and Plans
   Background:
       Want to produce a reference Java messaging system – MPJ.
       Aamir started working on the effort in Nov 2003.
       Working with Bryan Carpenter – now at OMII in Southampton.
   MPJ – layered architecture:
       Message passing API – similar to MPI.
       Secure runtime infrastructure.
   Current status:
       NIO device – complete,
       Sub-set of API,
       Basic infrastructure – embryonic.
   Issues:
       Technical – issues with NIO,
       Moving goal posts – emphasis on API/Infrastructure.
   Future:
       Complete reference implementation,
       Java MP/SM programming framework using MPJ/JOMP.
28 October, 2004           Mark.Baker@computer.org                   8
    Resource Monitoring
   Progress:
       GridRM framework has made significant progress
        over the last year:
            Consists of two layers – local and global (linked via
             jGMA).
       Deployed GridRM over international testbed.
       On-line demos - Cluster 2003/Taiwan/AHM
       Garry has submitted his PhD dissertation,
       RM review for Core Technologies book.




28 October, 2004          Mark.Baker@computer.org                9
  Resource Monitoring




   SDSC




28 October, 2004   Mark.Baker@computer.org   10
    Resource Monitoring
   Future:
      Garry – from 8
                      th Nov - will take up RA post,

      NeSC SC04 demo,

      SPE journal paper from thesis!

      Basis for e-Science proposal,

      Development:

           Enhance/extend testbed,
           More drivers,
           Installation/packaging,
           Sub-projects:
                  UG working on Portal,
                  Afrasyab Bashir – schedulers + broker.


28 October, 2004            Mark.Baker@computer.org         11
    JEM
   JEM (recap) consists of:
       Software modules that provide OS and other services,
       A mechanism to put them together and run them on the
        designated small device.
   Core modules are complete; JEM can run simple
    services on a VM on Linux or Myrinet NICs.
   The JEM project was completed in January 2003:
       Hong submitted and successfully defended his PhD thesis.
   JEM now in limbo, needs:
       Hardening,
       Soak testing,
       Packaging and release – plus decent Web pages!
       Further development – like a Java VM, modules,
       Papers…

28 October, 2004        Mark.Baker@computer.org                12
    UISB - #1
   The UISB project is developing technologies to
    interact, search and download Information Services’
    data, e.g. UDDI, LDAP, Jini LUS.
   Underlying idea:
       Search IS and retrieve all manner of high and low-level
        information that can be examined and potentially used by a
        developer:
           Textual – human readable view,

           Technical – so can potentially create client to interact

            with the “found” service.
   Technologies:
       Using Eclipse as the development/deployment platform.
       Semantic Web technologies:
          RDF as unified data store – so can join disparate info.

          XML/XLST/Xpath/Xquery/etc…


28 October, 2004         Mark.Baker@computer.org                 13
    UISB - #2
   Progress:
       Slowed by Eclipse development environment!
       Interaction with UDDI servers:
          For details see Rich’s talk.

       Looked at MDS3/4 – problems:
          Lack of documentation,

          Changing APIs (CoG),

          Unstable code.

       Currently looking at Jini (Java objects).
   Near future:
       Break out components from Eclipse,
          Look at using the RDF store with Mark H’s project,

          Integrate components into a Web portal using Portlet

            technologies.
       Rich giving a talk in Italy in December.



28 October, 2004          Mark.Baker@computer.org                 14
    jGMA
   Wanted a Java-based, lightweight!, easy to use,
    extensible version of GMA for GridRM.
   Was originally called JavaGMA, now jGMA.
   Testing and benchmarking over first half of 2004.
   Paper accepted and presented:
       At UK LUG in Bournemouth – January 2004,
       At AHM, Nottingham – August 2004.
   Started of with blocking/non-blocking API – first
    alpha (binary) release in June.
   Refactored in early summer:
       Now non-blocking API:
          Updated end-point addressing.

       Blocking API as add on later.

28 October, 2004        Mark.Baker@computer.org         15
    jGMA
   Progress:
       Non-blocking jGMA API has been thoroughly tested,
       Leasing, partially implemented (found very important),
       Demo for SC 04 – looks neat.
       Working on a scalable virtual registry:
          Based on persistent XML/SQL/text databases/files,

          Being designed – Mat will describe this work in detail in

            his seminar.
       Further benchmarking – compare with PyGMA/R-GMA.
       Full release.
       Full integration within GridRM,
       Start actively using it in other systems – on-line Games!



28 October, 2004         Mark.Baker@computer.org                  16
    OGSA-DAI #1
   Initiated within the OGSA Testbed project.
   Working with Southampton on GITS for OGSA-
    based systems:
       Perl scripts to examine and show the status of Grid
        services.
   Interested in OGSA-DAI as a pervasive technology
    for integrating relational databases.
   Decided to build a prototype system that pushes
    GITS data into a database that can then be
    searched and manipulated using an OGSA-DAI Grid
    Service and viewed via a Web browser:
       Helen is working on this project.


28 October, 2004         Mark.Baker@computer.org              17
    OGSA-DAI #2
   Progress:
       Perl script to parse GITS – completed,
       Poster in AHM,
       A number of technical problems:
          OGSA-DAI not rock solid yet,

          Interaction with mySQL and Xindice problematic.

          Starting to overcome hurdles.

       First prototype expected to be up and running by the end
        of November!
   Initiate a discussion with Bob N. to talk about using
    OGSA-DAI with SDSS – look at intelligent and
    efficient search mechanisms on large distributed
    databases:
       Research and develop a number of tools and techniques.

28 October, 2004        Mark.Baker@computer.org                  18
  SDSS




28 October, 2004   Mark.Baker@computer.org   19
    Other Projects
   Grid Performance:
       International effort:
            JISC funds to help with the workshop and pay for
             a report.
            Workshop held at Imperial College in May.
            Final report submitted to JISC in September 04.
            Potential follow up events – maybe NeSC!
   Portal development (Hong):
       Part of a JISC proposal to develop a bunch of
        services for Sakai.
       GridSphere being used for the liquid Crystal work
        (OGSA testbed).


28 October, 2004        Mark.Baker@computer.org            20
  Portal Effort




28 October, 2004   Mark.Baker@computer.org   21
    Performance Metrics
   Need to look at what are the measures of
    success for the group to be able to assess
    the potential of the DSG.
   The RAE uses:
       PhD completions,
       Income,
       Papers,
       Esteem/activities.
   Are these good metrics?


28 October, 2004     Mark.Baker@computer.org     22
  Growth of DSG
                                                   Aamir
                                          Mat

                                   Rich

                            Hong
                   Garry
     8
     7
     6
                                                                      + Helen + Mark H.
         Members




     5
     4
     3
     2
     1


              98       99     00   01       02         03   04   05    06
                                                Year



28 October, 2004                    Mark.Baker@computer.org                               23
         DSG Papers (per year)
                             X - Conferences + workshop
                             X - Int. Journal
5


4




2


1




    98       99         00     01        02            03   04   05
                             Year
     28 October, 2004        Mark.Baker@computer.org              24
    Funding proposals
   2002 – wrote/contributed to 7+ proposals:
       2 funded!
   2003 - wrote/contributed to 8+ proposals:
       2 funded!
   2004 - wrote/contributed to 5 proposals:
       JISC JCSR VRE – PD for two years, with
        Daresbury, Lancaster, Oxford.
       Basic technology programme - Paradigm Shifted
        Virtual Screening – PD for 3+ year, with Bio.
       Core e-Science – Grid performance modelling –
        PhD for three year, with Warwick and
        Nottingham.

28 October, 2004     Mark.Baker@computer.org        25
           DSG Income
50
                                 Myri




                                                                       EPSRC – E-science - 80K
                                                  EPSRC
                                                   45K
40




                                                                                                 JISC – VRE - 80K

                                                                                                                    AstroGrid – 100K

                                                                                                                                       With Bio – 100K
                                 CLRC
                                 14K



                         DERA
                          36K
20
                                 DERA                       JISC
     NHSE                         36K                       JCSR
                                                   IBM
10    5K                                           14K      20K
     TOPIC   TOPIC       TOPIC             CLRC
      10K     10K         10K              10K     CLRC         CLRC                                     CLRC
                                                    4K           4K                                       4K

      98      99         00      01         02      03          04                                             05

                                  Year
      28 October, 2004                Mark.Baker@computer.org                                                                                            26
    DSG Membership
   MAB – staff.
   Garry:
       Submitted PhD Oct 2004,
       RA from Nov 8th 2004,
       Take AstroGrid post in Feb 2005.
   Hong:
       Submitted and defended PhD Feb 2004,
       2-year RA post:
          OGSA Testbed RA (1 year), + Faculty RA (1 year).


   Rich - Faculty PhD bursary (1 year to go).
   Mat - DTA PhD bursary (2 years to go).
   Aamir - DSG PhD bursary (2 years to go).
   Helen – CSSE bursary (2 years to go).

28 October, 2004        Mark.Baker@computer.org               27
    Esteem (reflects on the group)
   TCSC (co-chair) and TCPP (exeCom).
   Visiting chair – Westminster/Reading.
   Steering Committee of Cluster/CCGrid/Grid.
   DSO section editor.
   Invited talks.
   Grid Book – early 2005.
   Reviewing (books, journals, conferences).
   Membership of conference PCs.
   Organisation of events.
   NeSC events.
   Rapporteur for IST CFP, Nov 2003.
   Hits on, and prize for Web site…

28 October, 2004   Mark.Baker@computer.org       28
  DSG Presence




28 October, 2004   Mark.Baker@computer.org   29
  DSG Activities




28 October, 2004   Mark.Baker@computer.org   30
    DSG Activities
   ICCC
       Melbourne, Australia, Dec 1999.
       Chemnitz, Germany, Nov 2000.
       Newport Beach, California, Oct 2001.
       Chicago, Sep 2002.
       Hong Kong, Dec 2003
       San Diego, Sep 2004
       Boston, Sep 2005
       Barcelona Sep 2006
       Texas 2007
   CCGrid
       Berlin, Germany, May 2002, Tokyo, Japan, May 2003, Chicago,
        April 2004, and Cardiff 2005
   Grid (with SC)
       Denver 2001, Baltimore 2002, Phoenix 2003, Pittsburgh 2004,
       With SC in 2005 too.
28 October, 2004          Mark.Baker@computer.org                     31
    DSG Metrics
   PhD completions:
       Hong - February
       Garry – November!
   Papers – need to produce at least one conference
    paper per year and one journal paper per PhD
    completion.
   Income – average, hopeful of success over coming
    year.
   Esteem – group is well know in the US and has
    started to establish a better reputation in the UK
    and EU.


28 October, 2004       Mark.Baker@computer.org         32
    SWOT Analysis
   Strengths:
       Emerging national recognition – (Testbed, AHM, SC demos),
       Very good relationships with Daresbury and via OGSA
        Testbed.
       Partially grown group – via visitor and CSSE,
       Semi-autonomous (good and bad).
       Group has extensive and complimentary skills,
       Continue to have greater interaction with the ICG and
        growing these links – especially now Bob’s around.
       Well connected, building even stronger links with EU and UK
        groups.
       Knowledge in depth Java technologies, gather experience
        with WS, and various W3C XML technologies,
       Profile via TCSC, DSO, and DSG Web presence,
       Undertaking synergetic projects which leads to good group
        focus.
28 October, 2004        Mark.Baker@computer.org                33
    SWOT Analysis
   Weaknesses:
       UoP CS research is still low-profile, making it doubly
        hard to get funds.
       Still no internal UoP staff help with getting funding.
       Still trying to get other UoP staff involved in DSG
        research, potential with ICG and CSSE PhD student.
       Administration and administrative effort is
        challenging and time consuming, Chris Duncan is
        helping us no end, but it still sucks a lot time.
       VERY Ham-strung having to work through ISO, e.g.
        fire walls, general attitude, etc.
       Still little industrial and/or commercial connections.
       May be losing Hong – hard to replace!
       Semi autonomous – justify our existence, not a very
        large umbrella.
28 October, 2004       Mark.Baker@computer.org               34
    SWOT Analysis
   Opportunities:
       OGSA Testbed – Great opportunity which has long term
       TCSC, educational and other collaborative changes,
       Mark Holliday – various openings,
       Building up links with Biology (Martyn Ford – BTP),
       Strong national links:
          Daresbury, Lancaster, Westminster, Reading,

           Southampton, and Manchester.
       Grid book – funding for companion Web site and
        development of material.
       Getting more involved in the e-Science programme,
       Semi-autonomous,
       With ICG we have the best research culture in the UoP!
       IEEE CS –TCSC/DSO…


28 October, 2004        Mark.Baker@computer.org                  35
    SWOT Analysis
   Threats:
       FUNDING…
       Involvement with GPN,
       Recruitment of students, RAs and more senior staff.
       Semi-independence via the ICG!!
       Internal faculty funding – will our funds be there in five
        years!
       Getting bogged down in teaching and administration – a
        large overhead is associated.
       Making sure DSG income gets tagged for the DSG research.
       Not retaining good staff!… losing staff!




28 October, 2004        Mark.Baker@computer.org               36
    SWOT Conclusions.
   As usual interesting comparing this SWOT with that of last
    year.
   Some fears regarding threats and weakness have happened:
       Group has not grown – limited funding, but had additional input
        from Helen and Mark H.
       Teaching and university processes are sinks of time and effort!
   With respect to our strengths and opportunities we seem to be
    taking advantage of them:
       Our reputation, skills and presence is helping us a lot,
       Optimistic about funding proposals.
   I still think our perceived strengths and opportunities out-
    weigh the weaknesses and threats, but we still need to be
    careful!
   The biggest threats and weakness are on the still on funding
    side (internal and external).
   Our strengths are focussed on applied research, useful skills,
    good interactions, and lots of enthusiasm!
28 October, 2004           Mark.Baker@computer.org                    37
    Summary
   I think we’ve had a reasonable year.
   We have managed to:
       Enhance our reputation,
       Produce some good ideas,
       Continue to get some income,
       Maintaining productive research culture.
       Widening our experience and collaborations.
   However:
       No new students this year,
       Limited new funding,
       Lots of doing, probably not enough papers.
   Congrats to Hong – successful defence of PhD.
   Good luck to Garry submitted PhD, viva soon!
   I told you we were getting better at Web design.
28 October, 2004        Mark.Baker@computer.org        38
    Five Years Hence (2002 view)
   Group size of ten plus – permanent staff, post docs
    and post grads.
       Coherent and strongly focused research.
       Independent group working in a strong research culture.
   Broken down the external view of research at UoP.
   Internationally recognised in a number of related
    fields:
       Wide-area resource monitoring,
       Java micro-kernel work,
       Java-based Grid technologies.
          Deployed and regularly downloaded software.


   Papers – 30+ conference and journal papers per year.
   Invites to collaborate/talk at international
    events/institutions.
28 October, 2004        Mark.Baker@computer.org                   39

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:2
posted:11/9/2011
language:English
pages:39