The e-Science Core Programme

The e-Science Technical Advisory Group

Minutes of the 18^th meeting

27 April 2005

DTI, 151 Buckingham Palace Road, London

(Agenda)

Present:

Prof. Tony Hey (Director, Core Programme, DTI/EPSRC) Chairman
Dr Jamil Appa (BAE Systems Advanced Technology Centre)
Prof. Malcolm Atkinson (NeSC)
Dr Graham Cameron (EBI)
Prof. Jon Crowcroft (Cambridge University)
Prof. Peter Dew (Leeds University
Dr Alistair Dunlop (OMII)
Prof. Geoffrey Fox (University of Indiana)
Dr Paul Jeffreys (Oxford University Computing Services)
Linda McCormick (Glasgow University)
Prof. Ron Perrott (Queens University Belfast)
Prof. Nigel Shadbolt (Southampton University)
Prof. Paul Watson (Newcastle University)

In Attendance:

Deborah Miller (PPARC) Secretary
Audrey Canning (DTI)
Dr Jim Fleming (EPSRC)
Dr Neil Geddes (CLRC)
Dr Alan Robiette (JISC)
Dr Anne Trefethen (EPSRC)

By Invitation:

Dr Michael Jubb
Dr Malcolm Read (RLN)

Apologies:

Dr Mike Kirton (QinetiQ)
Dr Angela Sasse (UCL)
Prof. Ian Sommerville (Lancaster University)
Prof. David Williams (CERN)

 

Minutes for the 17^th meeting

1

The minutes were agreed as an accurate record of the meeting.

Actions and matters arising from the minutes

2

The following matters arising from the minutes were discussed:

a

Minute 2c: Linda McCormick to circulate details of the UCISA/JISC campus security website. Done

b

Minute 3c: Members to suggest names to brand the e-Science grid vision. Tony Hey said that the term “Plug and Play Composable Services” had been chosen.

c

Minute 3d: Jim Fleming to provide further information on the Sisters Project Done (Annexed to the minutes of that meeting).

d

Minute 3d: Tony Hey to present a report on the Sisters Project to the July 2005 TAG. In hand.  

e

Minute 3e: Malcolm Atkinson to consider holding a BoF session for the Sisters Projects at the All Hands Meeting.   Malcolm Atkinson said that all the BoF slots were full, so something informal would be arranged instead.

f

Minute 3h: Malcolm Atkinson to brief Peter Henderson on his role of liaising with the GridCoord project. Malcolm Atkinson said what was really required was a UK consensus view on GridCoord and other similar activities. To achieve this, the eIRG required reforming to provide a consistent UK input. Tony Hey and Neil Geddes agreed to look at how this might be achieved. Action: Tony Hey and Neil Geddes

g

Minute 5b: Tony Hey to develop a strategy for engagement with China. Tony Hey reported that the MoU with China was about to be signed, and following that, OMII China would be set up. Two Chinese researchers were currently at the OMII and two further researchers would join OGSA-DAI in May. Geoffrey Fox said that it would be a good idea to set up a branch office in China to facilitate liaison with Chinese companies. However, it was noted that several other similar offices were being supported by universities, RDAs, the EU, HEFCE etc, so caution was required. It was agreed that this suggestion should be discussed more fully at the July meeting. Tony Hey would consider preparing a background paper, which would include a summary of existing offices. Action: Tony Hey Geoffrey Fox agreed to put his thoughts on paper and sent them to Tony Hey Action: Geoffrey Fox

h

Minute 7l: and 7m Deborah Miller to clarify DCC reporting lines with the e-Science Core Programme. Done. The Director of the DCC, Chris Rusbridge, would now formally report to TAG and had been invited to present a report to TAG at its July meeting. Malcolm would continue to attend TAG meetings and provide informal updates.

i

Minute 8c: Angela Sasse to present a paper on the usability aspects of security. This would be presented at the July meeting.

j

Minute 8f: Malcolm Atkinson to flag up the presentation of the results of the JISC security evaluations as a future e-Science event at NeSC. Ongoing. Malcolm was discussing this with Anne Trefethen and Alan Robiette's team.

SR2004 Outcome

3

Tony Hey presented his report (.doc .pdf ) The following issues arose:

a

Most of the SR2004 money had already been allocated, so future funding for e-Science was tight. 

b

EPSRC was hoping to engage with other Programmes within EPSRC to gear up e-Science funding.

c

There was no common Research Council policy for a ring-fenced e-Science budget and statements on future e-Science activities had not yet been received from ESRC, BBSRC and NERC. Concern was expressed that many of the Research Councils were planning to embed e-Science activities within other programmes and activities and that this lack of “badging” would lead to a loss of a common vision, visibility and cross-Council collaboration.

d

Further discussion with the Research Councils was required to ensure that the badging was not lost.

Priorities for Phase 3 of the e-Science Core Programme

4

Anne Trefethen gave a presentation on future priorities for the e-Science Core Programme (.ppt ).  The following comments were made:

a

A sustainable model for spin-out was required. This was an issue common to JISC as well and had been the subject of a paper by Alan Robiette.

b

Future funding of the OMII, GOSC and DCC would be subject to review. Peter Dew suggested that they be asked to develop a business plan for sustainability.

c

Tony Hey would raise the issue of future engagement with the DTI, with the Chair of Technology Programme Strategy Board.

d

Malcolm Read said that JISC funding for e-Infrastructure was £10.6M over 3 years and there were 3 relevant sources of funding: a) national infrastructure e.g. JANET, but there were no earmarked funds for e-Science, b) the middleware area where there was significant funding, but any specialised needs of the e-Science community e.g. in terms of a separate authentication system, were outside the scope of JISC and c) JSCR, where funded activities would have to impact the whole research community. Possible areas of support would be the NGS, UK Light, OMII and training and advice. There was a need to put the support of e-Science in the context of other demands from JISC e.g. Digital Libraries also wanted an e-Infrastructure

e

Malcolm Read said that there was an “amber light” on capital funding of £5M per annum. This was short-term funding but would be appropriate for e-Science, if again, it was supporting the whole research community. A proposal for £10M for VRE was gong to the HEFCE Board in June.

f

Ron Perrott said that HPC would be integrated into the national e-Infrastructure as an affiliate to the NGS. Other Research Councils should be encouraged to integrate HPC in a similar way.

g

Tony Hey had been talking to EPSRC about the future management of the e-Science Programme. It was possible that it would be amalgamated with the HPC Programme.

h

Regarding integrated access to national data sets, Jon Crowcroft said that the e-Science Programme should be talking to companies with experience of real systems e.g. Google. He agreed to work with Tony Hey and Malcolm Read in developing ideas on how to take this forward. Action: Jon Crowcroft

i

Malcolm Read said that the issue of where data was held was important as some researchers did not have access to sufficient bandwidth. He was talking with the Research Councils about making their data resources available in the same common AA environment e.g. Shibboleth as the Athens replacement. JISC would like the Research Councils to create a culture where making data available to others was as important as publishing research papers. He would like to create a vision in e-Science of a national infrastructure for data content, but this needed to be taken forward strategically. Tony Hey agreed to raise these issues to the EU eIRG. Action: Tony Hey

j

The focus on a persistent infrastructure was good but needed to be put in an international context.

k

There would be a call for e-Science Centres and it was hoped to fund up to a maximum of 8 to10 centres.

m

Geoffrey Fox said that there should be an up-front statement on progress or lack of progress to date.

n

Sustainability was a key issue. Spin-off technologies would provide one form of sustainability. Linking in content would encourage a multidisciplinary approach. Jamil Appa was concerned about links with the DTI and the need to market e-Science within the industrial community. Audrey Canning said a description was required of the potential beneficiaries and what they would get out of the programme. For example, what was the benefit to the DTI?

o

An overarching vision for UK e-Science was required. Tony Hey agreed to circulate a paper on this for comment. Action: Tony Hey

Transition Plans

5

Tony Hey presented his paper (.doc .pdf) on how the present management structures of the e-Science Core Programme could evolve for Phase 3. The following issues arose:

a

Ron Perrott said that the HSC had not yet seen a copy of the paper, but he could see possible benefits of joint committees and cross-referencing of documents.

b

The paper need a preamble about what the management structure was trying to achieve plus metrics for success.

c

The Research Councils would not have individual membership of the eISC, as it would make it too large, but there was a need for all the Research Councils to have a stake in the management structure.

d

The e-STC was dominated by people already funded.

e

A rotating Chair for eSTC would result in the loss of continuity and a single point of contact. Tony Hey was asked to request a part-time Directorate to maintain this.

f

Members were asked to suggest alternative names for eSTC. Action: All

Grid Operations and Support Centre and the National Grid Service (GOSC&NGS)

6

Neil Geddes presented his report (.ppt) on the Grid Operations and Support Centre (GOSC) and the National Grid Service (NGS).  The following issues arose:

a

MTBF statistics were not collected as the contracts with the NGS did not require service levels. Jon Crowcroft said it was important to monitor log-on correlated failures.

b

The users had been polled for science papers and this would be repeated every six months.

c

It was planned that HECToR would be at least an affiliate with the NGS, but a partnership would be preferred.

Open Middleware Infrastructure Institute (OMII) and OGSA-DAI

7

Alistair Dunlop presented his report (.ppt) on the OMII.  In discussion the following issues arose:

a

Statistics were not yet available on how many users migrate through the releases. Users were being polled on what they are using the down loads for and this would be reported at the next meeting.

b

There were very few users from the EU. After the release of OMII-2 there would be compatibility with GT4 and OGSA-DAI, and OMII expected to see increased user take up.

c

OMII intended to be an OGSA-compatible platform and so wanted GGF to approve OGSA.

OGSA-DAI Update

8

Malcolm Atkinson gave and update on OGSA-DAI. The following items were reported:

a

The OGSA-DAI user base was growing steadily.

b

A 3^rd User Group meeting would be held at NeSC on 1 June. It was important for UK users to say what they wanted from OGSA-DAI: US users were already vociferous in their requirements.

c

OGSA-DAI's experiences with GT4 were fairly good, but there was a challenge in the clarity of naming where services, data, information could be uniquely identified.

d

Students were encouraged to attend the International Summer School on Grid Computing to be held in Naples, in July.

Research Libraries Network

9

Michael Jubb gave a presentation on the work of Research Libraries Network (ppt.) The following issues were raised during discussion:

a

The focus of the RLN was much wider than just libraries and was concerned about building a sustainable, national research information resource. Its scope covered hard copy and well as digital information sources

b

In e-Science search and delivery was not a practical system for large datasets; some local computation was required.

c

The RLN had a wide-ranging scope with many interfaces with other bodies and there was a need to broker discussion between these bodies rather than create further competition for funding. Mapping was required of the various initiatives and gaps.

Security Update

10

Anne Trefethen presented an update on security activities (.pdf .doc). The following points were raised during discussion:

a

Athens was now providing authentication resources for 3M users. It was an incomplete solution, but robust.

b

The security usability audit would be presented to the July TAG meeting. Some specific recommendations were coming out of this work e.g. a portal for security information perhaps linked with JISC information services or the NGS host portal. The audit would show what users were looking for.

c

Researchers and administrators were saying that they found the whole issue difficult to understand and needed to be guided on what to do through “cookbooks” and a longer-term education programme.

d

There was a need for documentation that addressed system-wide security issues such as denial of service attacks.

DTI Technology Programme Update

11

Audrey Canning gave an oral report. The following items were reported:

a

7 projects worth a total of £9M were funded in the April 2004 call under the Inter Enterprise Computing (IEC) theme. A further £1M was awarded for two Knowledge Transfer Networks in the IEC area in which the NeSC was heavily involved.

b

The Nov 2004 call had passed through the 1^st round. 976 bids had been received against the 10 priority areas. 389 had been invited to make full bids. Demand was £225M against the £100M available so success rates at this round were expected to be approx 35%.

c

The April 2005 call had a limited focus on e-Science community interests. Only the complex systems theme was of possible interest.

AHM 2005 Update

12

Ron Perrott said that the arrangements for the AHM in September were on track. Tony Hey would be speaking to the RC CEO's about attending the meeting and their roles. An announcement would be made on 13^th May of papers accepted for the meeting. About the same number of papers were submitted as last year.

SC2005 Update

13

Anne Trefethen and Jim Fleming were discussing with John Gordon and Dave Berry about a call for demonstrations at SC2005. The stand would be redesigned. One project had been funded under the EPSRC call for Data Intensive Science Demonstrator projects.

Director's Report

14

Tony Hey presented his Director's report (.ppt). The following items arose during discussion:

a

The US was developing a coordinated e-Science policy.

b

Tony Hey agreed to circulate the website of the recent `Berlin-3' Open Access Workshop held at Southampton. (It is http://www.eprints.org/berlin3/program.html) Action: Tony Hey

AOB

15

The next meeting would be held on 19^th July 2005.

Action List:

2f:

Tony Hey and Neil Geddes to look into how UK input to GridCoord might be achieved.

2g:

Tony Hey to consider preparing a background paper on engagement with China. Geoffrey Fox to send his thoughts on this issue to Tony Hey.

4h:

Jon Crowcroft to work with Tony Hey and Malcolm Read in developing ideas for integrated access to national data sets.

4i:

Tony Hey to raise the issue of a national infrastructure for data content with the EU eIRG.

4o:

Tony Hey to prepare to paper on the vision for e-Science.

5f:

All to suggest alternative names for the proposed e Science Town Council.

14b:

Tony Hey to provide the web address of the `Berlin-3' Open Access Workshop.