I visited Manchester University England with Dr Richard Hughes-Jones and Professor Robin Marshall. We ported UDPmon to SLAC and Caltech. We successfully executed it between hosts at SLAC and from Caltech to Manchester were able to achieve UDP throughput (no packet loss) of up to 350Mbits/s. Our efforts to use it to transfer data between SLAC and hosts at Caltech and Manchester were frustrated by a problem that appears to be related to SLAC's security measures. We arranged a follow up visit by Richard to SLAC coordinated with the ICFA/SCIC meeting at SLAC at the end of May. I was able to demonstrate to Richard the web-site for the SLAC led Internet End-to-end Performance Measurement (IEPM) Band Width (BW) project, and we started setting up a machine with a Gbps interface at Manchester located close to the Manchester GigaPoP to act as an IEPM-BW responder. At a later stage we agreed to port the monitoring, analysis and reporting package to Manchester for use by the European Data Grid community.
We also discussed the status of the ESnet QoS project and how SLAC might be involved in a UK QoS project.
Over dinner Robin Marshall, Richard Hughes-Jones and I discussed a possible leave of absence at Manchester for Warren Matthews of SLAC. This should facilitate inter-working between SLAC and PPDG and Manchester, the European Datagrid and UK network monitoring projects.
In Bucharest, we spent the first day making presentations to an audience of about 200 people who represented research institutes, universities, and two ministries (communication and information technology, education and research). Following presentations from the ministers, presentations were made by the invited visitors who included: Harvey Newman, Ian Foster, Fabrizio Gagliardi, David Williams, Mike Wilde, Olivier Martin, Iosif Legrand, Colin Buffam and myself. The second day was more of a round table workshop to define a project in the framework of the Romanian government. This will be called RoGrid and will be initiated on April 22nd '02. It will be put in place as a legal entity by June 1st '02 and will cover research institutes, universities and the private sector. RoGrid will represent Romania in grid activities.
Coordination of activities between the SLAC led IEPM-BW project and European Datagrid network monitoring work. Updating on QoS plans, and iGrid2002. Learning of Romanian computing activities, and keen Grid interests. Initiating a collaboration between the SLAC led DoE funded IEPM project and Romanian universities on Internet measurements. Laying ground work for future collaborations on Grid related projects.
At Manchester University I contacted Dr. Richard Hughes-Jones and Professor Robin Marshall.
In Romania I met Olivier Martin, Fabrizio Gagliardi and David Williams of CERN, Harvey Newman and Iosif Legrand of Caltech, Mike Wilde and Ian Foster of ANL. I also met Dan Nica minister of Communications and Information Technology and Adrian Campurean the secretary of state for research and parliament relations and their staffs, prof. dr. Doina Banciu of the National Institute for R&D in Informatics, prof. dr. Valentin Cristea Vice Dean of the Politehnica university of Bucharest.
In London I contacted Professor Peter Clarke of University College London.
The first meeting was at the Victoria University of Manchester Physics
The meeting with Peter Clarke was at Balliol College Oxford, UK.
The "Potential of Grid and Distributed Computing Activities in Romania" was held in the Parliament Palace, Hall Bratianu. The following day we visited the National Institute for Research & Development of Informatics (ICI (8-10 Averescu Avenue)), and the Polytehnica University both in Bucharest Romania. In the evenings we had dinner with Dan Nica minister of Communications and Information Technology and Adrian Campurean the secretary of state for research and parliament relations some of their staff and some University professors.
Dr Richard Hughes-Jones, Professor Robin Marshall
I visited Manchester University England with Dr Richard Hughes-Jones and Professor Robin Marshall. Richard is an International Committee on Future Accelerators (ICFA) Standing Committee on Inter-regional Connectivity (SCIC), an active member of the European Data Grid (EDG) Working Party 7 (WP7), collaborates with the IEPM project at SLAC and is the author of the UDPMon package which SLAC is evaluating. While I was there we ported UDPmon to SLAC and Caltech. We successfully executed it between hosts at SLAC and from Caltech to Manchester were able to achieve UDP throughput (no packet loss) of up to 350Mbits/s. Our efforts to use it to transfer data between SLAC and hosts at Caltech and Manchester were frustrated by a problem that appears to be related to SLAC's security measures. We arranged a follow up visit by Richard to SLAC coordinated with the ICFA/SCIC meeting at SLAC at the end of May. I was able to demonstrate to Richard the web-site for the SLAC led Internet End-to-end Performance Measurement (IEPM) Band Width (BW) project, and we started setting up a machine with a Gbps interface at Manchester located close to the Manchester GigaPoP to act as an IEPM-BW responder. At a later stage we agreed to port the monitoring, analysis and reporting package to Manchester for use by the European Data Grid community.
DT WP2 wants to look at QBSS since it appears to be an obvious candidate. They will try it across the UK testbed (between Manchester, Rutherford and UCL) and the DataTag development network. After running testbed the idea is to extend the test network to true end to end performance. E.g. at the end of the UK project we might want to do tests between SLAC and DL, using class C networks at each with the appropriate announcements. Timescales on UK project are to have access to UKERNA development network in summer '02 and so will run a set of tests then, including, maybe a year off to doing end-to-end to SLAC.
Action items include:
RHJ get host for IEPM-BW to monitor at RAL
SLAC get the NWS prediction code and try it out.
SLAC look at RHJs query code modification to add column selection and incorporate into PingER if appropriate.
SLAC look at LDAP code from WP7
Les investigate getting a copy of the SC2001 real time PingER/iperf demo for iGrid2002.
We will try and make progress on these action items before the next meeting in Manchester.
Peter is an organizer of the Global Grid Forum (GGF). There is a well funded, high-profile project called Managed Bandwidth Next Generation (MBNG) that has as one of its goals to demonstrate using Differentiated Services (DS) between CERN and StarLight in Chicago. They hope to have the basic infrastructure in place in the UK by September '02. They will be using Cisco 6500s. A next step would be to extend to other sites in Europe and the US. To enable this contacts with Abilene and/or ESnet staff will be need as well as Collaborators in the US. Les will assist in making these contacts, in particular with linking up with the ESnet/SLAC QoS project.
Peter is deputy manager of Working Party 7 (WP7). Yee of UCL is working on monitoring for Grid activities. There is overlap between the SLAC IEPM-BW project and Yee's work. One of the issues we both recognized is how to allow for publishing and subscription to the data. Yee has been looking at some web services and XML to assist in this.We need to get together to align our efforts. To facilitate this Peter plans to arrange a visit by Yee to SLAC to share ideas, understand the two approaches. Les will set up an H.323 video conference with Connie, Warren & Les at SLAC and Peter and Yee at UCL to present our goals, methodologies and results followed by discussion on where we should focus our efforts.
We discussed where Warren Matthews might fit in in the UK for a year's leave of absence from SLAC. UCL would be one natural place since it has a networking group consisting of five people, and there may be funding from the European DataGrid.
This was a conference arranged by the Romanian Ministry of Communications and Information Technology and the ministry of Education and Research. After a day of presentations mainly from invited foreign experts in the Grid field we met the next day to formulate a plan of action. Five points were set out to be addressed: define the project in the framework of the Romanian government, define cooperations with outside of Romania, define the infrastructure needed, define the bandwidth needed.
The RoGrid project was defined with a starting day of April 22 '02 (the following day). It will be publicized and be in place as a legal entity by June 1st '02. It will be a consortium covering research institutes, universities and private sector. It will represent Romania in Grid activities. For more on this see the report in the Romanian English language daily "Nine O'Clock" April 22, 2002.
The requirements to support RoGrid were identified as:
Grid software development and deployment. Need a computing system that is modern with a commitment to keep it up to date. Will need to be able to contribute something to the grid. Network infrastructure technologies separate thrust for transforming Romanian society, for research and education, integrated into the grid project, provide a road-map (optical, wavelength based) to bring Romania in coincidence with the state of the art. Distributed systems simulation design and optimization. Collaborative systems deploy scalable infrastructure, develop integrated grid-enables environments for a variety of disciplines.
Romania now has 155Mbps to GEANT as of 2 weeks ago. Need to increase to 2.5Gbps. Need to extend to universities & research institutes. Need to work with EC (Brussels) to improve communications infrastructures. Need to catch up to today in a 2 year time frame, i.e. today 2.5Gbps is the standard for backbones, so Romania needs to get there within 2 years.
Will build connectivity to 7 major universities, and will merge the existing research networks.
I gave a presentation on Worldiwde Network performance and Monitoring.
We (Harvey Newman, Iosif Legrand of Caltech and I) met with Profs Valentin Cristea, Vasile Buzuloiu and Nicolae Tapus of the PoliTehnica university. We talked about possible collaborations. The following areas were identified as being of joint interest from the US end:
The Politehnica has expertise has skills in simulation, monitoring and databases. Might be important to focus on one of the areas.
Needs to be a collaboration, i.e. both sides benefit. Concerns about one-way brain drains of students, some will go, but need to ensure that people can collaborate effectively from Romania, i.e. needs good network infrastructure for collaboratory, data sharing, access etc.
Next step is to communicate that we have identified a few initial focus areas of common interest. That even for these modest collaborations will need improved networking.
I met with some students and in particular with two who will probably be the people who will install an IEPM-PingER monitoring site at the Politehnica.