AG-SAG - Full Meeting #7

---

---

Date and Location: Centre for Astrophysics Research, University of Hertfordshire, 08 Sep 2005, 10.00-16.00

Supporting Paperwork:

---

Agenda

Present:

• SilviaDalla (SD) (AstroGrid) (y)
• EduardoGonzalez (EG) (Astrogrid) (y)
• AndrewJaffe (AJ) (y)
• NialTanvir (NT) (y)
• JonathanTedds (JT) (Astrogrid) (y)
• NicholasWalton (NW) (Astrogrid) (y)
• JeremyYates (JY) (y)
• Seb Oliver (SO) (y)
• GuyRixon (GR) (Astrogrid) (y)
• BillChaplin (BC) (y)
• RichardMcMahon (RM) (Astrogrid) (y)

Apologies

• Carlton Baugh (CB)
• Lyndsay Fletcher (LF)
• Danny Lennon (DL)
• MikePinnock (MP)
• Tom Marsh
• Steve Maddox
• AnitaRichards

Notes of Meeting

Review of Feb 05 meeting minutes (SAG6) and actions

Previous Actions

ACTION FM6-1:AR: number of AG users

As of Sep 7 2005 there are 106 registered users of the AstroGrid system.

AJ: Some concerns have been raised regarding load to the system. JY: UCL offered some machines - to be followed up.

ACTION-FM6-2:NW will provide a list of current applications on the priority list for SAG members to check.

• The wish list of applications for inclusion into Astrogrid is at ApplicationWish
• The list of currently deployed server side applications is noted on the Deploy.DeployedComponents page
• The list of current client side apps available through AstroGrid is maintained at HelperApps

NW: Most of these applications are linked to use cases. Some of these applications are quite compute intensive - good to run them on a grid infrastructure rather than on own PC.

ACTION-FM6-3:NW Glossary of terms for users needed

We keep an acronyms wiki page online at VO.AcronymsAndLinks. We should try and integrate this into the Glossary

ACTION-FM6-4:NW need form of words to acknowledge Astrogrid usage in refereed papers

A suggested form of words - based on those used by the NVO (see http://www.us-vo.org/ack.cfm) - could be:

'This research has made use of data obtained using, or software provided by, the UK's AstroGrid Virtual Observatory Project, which is funded by the Particle Physics and Astronomy Research Council and through the EU's Framework 6 programme.'

The minutes of the last meeting were approved.

Feedback from Astrogrid Oversight Committee Meeting, Feb 05 (Report) and progress towards the Oct 2005 AGOC meet

AJ: SAG needs to report to AG Oversight Commettee in October.

Astrogrid development through to NAM2006, release v1.1 and beyond (see slides NW)

Current release is V1.1.

Release 1.2 will come out at the end of 2005 - will have ColourCutter and AstroScope.

Discussion of Cycle 2 plan (till end of 2005).

AJ: Regarding the new applications listed in the wish list, running some of these applications is in many cases quite complicated. They have complicated input files etc.

JY: It would be best to make library of solutions available, rather than the applications themselves.

AJ: I am concerned about the fact that this making applications available via the grid will drain resources form the main focus of Astrogrid - making the data available.

SO: The applications that you need to focus on are the ones that people do not currently run on their desktop, for which they really need more compute resources.

NW: We do not aim to run big simulations, just smallish applications that are useful in relation to observational data.

AJ: There will be a call for HPC applications (deadline probably April 2006) - possibly Astrogrid should put in an application to ask for funds to establish a large compute resource.

Demo of release v1.0 and v1.1 (Open to ALL)

Well attended by some 15 Herts staff and students. NW, SD and EGS gave a demo of the AG system, redshift maker and solar movie maker in real time.

Some feedback to demo: S-Extractor works for the optical. It would be useful to have a similar tool for radio data

Technical update (inc IVOA issues)

SO: How do future surveys make sure that their data will be included in AstroScope?

NW: It would be good to have a record of future projects, like surveys etc, in the registry.

GR: It might be possible to develop a tool that makes integration of datasets into Astrogrid in future easier.

Extending Astrogrid to STP - recent results

SD: mentioned the STP wok that followed on from the JUl 05 Workshop, and in particular the CTIP model, and the connection to STP datasets held at the UKSSDC in RAL.

SD discussed progress on the STP and a new solar science case relating flare production to active regions.

Update on Science priorities for AG2

NW: New Science Services will be added, and the Redshift Maker will be improved.

AJ: In addition to the Science Services, examples of what can be done with AstroGrid would be very useful to new users. Not necessarily something very polished and that will work in all cases.

BC: This is a way in which the SAG can help Astrogrid, by developing examples of usage.

JY: Is Astrogrid at a stage where we can think of Msc and Phd projects to develop tools to be made available via AG?

BC: Is there a way for people to share what they have done with Astrogrid. How can we facilitate this?

NW: People that attended the Astrogrid workshop attached workflows to wiki pages.

AJ: It would be useful to have a single wiki page that points to workflows that can be used and tried.

JY: Developing the Gould Belt science case showed quite a few issues where improvements can be made. For example, make available a radiative trasfer code.

Applications update c.f priority list

AJ: A key question is how to make applications interact with votables and files in Vospace. So that you can plugin objects from vospace into your favourite applications.

NW: Soon you will be able to use Python to do scripts. Also there are libraries so that you can talk to the ACR by using Python.

SO: Many people use IDL and it would be good to have an interface between IDL and VOspace. The same for other common languages.

SO: Cosmology applications would be very good - luminosities, NEDWRIGTH cosmology calculator, etc.

SD: An application for visualisation of time-series and also for time-series analysis. NT: Possibly have a look at time-series application by Scargill.

SO: Unit conversion application.

Science input to the VOTECH stage 2 planning

See above.

Increasing the use of the Astrogrid system

Feedback from AG/SAG jul 05 workshop - see AgWorkshopJul05

JY: Atmospheric Science people at UCL will develop CTIP model work started at Astrogrid workshop.

SO: People from Sussex have come back pretty enthusiastic from workshop and plan to do more work. Also organise a workshop at Sussex, possibly have Eduardo join in.

NW: There will be a solar workshop at MSSL soon. I would like to have similar workshops in various groups that expressed an interest in Astrogrid. Have a couple of people from Astrogrid at these workshops.

SO: Alternatively try and group people by interest, eg all SWIRE people.

NW: There needs to be some funding for departments that now start deploying Astrogrid services. There are now operational issues related to running the Astrogrid system.

JT: We should invest also in making visualisation better and easier so that people have that as an extra reason to start using the system.

SO: It would be good to focus on PhD students and postdocs. Possibly take part in PPARC Summer Schools.

Development of use cases AG with SAG

JY: Report on GouldBelt science case. 2MASS and IRAS data selected and processed, and colour-colour diagrams produced. The idea was to compare them with diagrams obtained from models - very difficult to do with Topcat ! Also, we need radiated transfer code made available.

Future Astrogrid meetings and Workshops

Next Meeting: Thursday 2nd February 2006 (TBC), Sussex University.

AOB

---

Travel and Accomodation

The meeting will be held at the Centre for Astrophysics Research, Univ of Hertfordshire - see the web pages at http://strc.herts.ac.uk/astro/. Info on how to get to Herts is given at http://strc.herts.ac.uk/astro/visitors.html.

• the location is room E115 in the "Science and Technology Research Institute" aka STRI. It's only a couple of mins walk from the main entrance of the College Lane campus. Getting in to STRI does involve a swipe card door, so probably best to try to ring me from reception if you don't already know your way in.

Getting to University

• by rail, just need to get taxi to College Lane campus
• by road, need junction 3 off A1(M), visitors car park is to left (signed) off the main drive a couple of hundred metres from the entrance. Once you park you'd need to come to main entrance to sign up for visitors car parking badge. If people let me know (by Tues 5 Sep) that they are coming by car I may be able to get a labelled space - although that sometimes works and sometimes doesn't! (If visitors car park is full, then best to carry on to Car Park 9, which is a little further away)
• Some maps: http://perseus.herts.ac.uk/uhinfo/extrel/info/maps/maps_home.cfm

Parking

As you come into the Uni (which is just off jct 3), there's a visitors car park on your left after about 200m. It's fairly clearly signposted. If full, carry on to car park 9 (signed "conferences and events"), which should certainly have spaces, but is slightly further away.

In either case you need then to go to the main reception at the front of the uni and ask for a visitors parking permit - just sign in that you are visiting me. Permit then needs to be displayed in car, I'm afraid - hope its not raining.

Wireless and Network Connections

No wireless, there will be ethernet connectivity available (provided via an Astrogrid router/hub)

-- NicholasWalton - 30 Aug 2005