[DAIS-WG] DAIS Interop Telcon Minutes

[Mario Antonioletti]

DAIS Interop Call - April 16th 2009
===================================

Present:
         Mario Antonioletti, EPCC
         Steve Lynden, AIST
         Giacinto Donvito, INFN
         Oscar Corcho, UPM
         Sunil Ahn, Kisti

Agenda
------

     - Agenda bashing
     - Process
       - Agree on:
             * binding document/literal or rpc/literal
       - Come up with a scenario and database to use
       - Putting up end points to test
       - Coming up with a test suite
       - Roles
       - Timelines

Actions
-------

[MAA] speak to the ByteIO folks to determine whether they had a
       rationale for using document/literal and establish whether there
       is a preference in OGF in general.

       Not done yet.

[MJJ] check whether changing the OGSA-DAI WS-DAIR implementation to use
       document/literal would entail a lot of work.

[MJJ] Put a version of the DAIX test-suite on the DAIS Grid Forge.

Completed Actions
----------------

[SA] Document the changes required to change the WSDL to work with
      gsoap for rpc/literal.

       Done. Information at:
       http://tinyurl.com/dafkde
       Under Binding Issues

[MAA] Publish the DAIS WSDL and XSD on GridForge.

       Done. WSDL & XSD at:
       http://tinyurl.com/chlxcn

[SA] Put up some information on the scenario/test suite used by AMGA
      to do their testing.

      Done. Info at:
      http://tinyurl.com/dafkde
      under Test Suite
---

Still need to agree on whether to use document/literal or rpc/literal.
To move forward on this need to determine:

  - how easy/difficult it will be for the current OGSA-DAI DAIR implementation
    to go from an rpc/literal to a document/literal implementation and
  - whether ByteIO had a rationale for using document/literal

In terms of having a test suite AMGA have developed SOAPUI which
encapsulates 30 or 40 SOAP messages that it will send to a WS-DAIR 
service and that tests the SOAP response.

Uses a very simple table with 4 attributes with 2 or 3 rows as the
underlying database data.

Giacinto would like to do some performance measurements too. Mario said
that this would be ok but not part of the interop process.  Also had a
discussion about dealing with complex SQL queries.  Giacinto was
wondering whether it was worth testing this too. Mario thought that
DAIS says nothing about the SQL sent to the underlying data
resource. This could be done but it would not really be part of the
interop - Giacinto could try it out if he wanted. Sunil mentioned that
the NIST SQL Test-Suite (http://www.itl.nist.gov/div897/ctg/sql_form.htm) 
could be used for this kind of testing as this had complex queries in
it.

AMGA have put up a document/literal end point (details published on
the wiki - http://tinyurl.com/chlxcn . It would take about 2 weeks for
the SOAPUI test suite to be ready.

In the meantime would try to determine whether OGSA-DAI can use
document/literal and whether ByteIO had a reason for using
document/literal in their interop.

Next call 8am UK time (GMT+1) April 27th 2009.

+-----------------------------------------------------------------------+
|Mario Antonioletti:EPCC,JCMB,The King's Buildings,Edinburgh EH9 3JZ.   |
|Tel:0131 650 5141|mario at epcc.ed.ac.uk|http://www.epcc.ed.ac.uk/~mario/ |
+-----------------------------------------------------------------------+
-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.