More on bulk throughput
Dr. R. Les Cottrell, MS 97, Stanford Linear Accelerator Center (SLAC), 2575 Sand Hill Road, Menlo Park, California 94025, <email@example.com>
The main part of the demonstration will be to show the achievable throughput of various applications from SC2002 to over 35 host in 8 countries. For this we will use real live production networks with no efforts to manually limit other traffic and no attempt to use MTUs > 1500Bytes. We also will demonstrate the effectiveness of QBone Scavenger Service (QBSS) in managing competing traffic flows and on the response time of lower volume interactive traffic on high performance links.
On the SC2002 floor we will have a 5-10 high performance Linux hosts each with two Gbit/sec network interfaces connected to a Cisco 65xx series Catalyst switch to be located in the SLAC/FNAL booth. We plan for the switch to have a 10Gbps and a two 1Gbps links to SCiNet and a built in router capability. The hosts will will run various high throughput applications including iperf and bbcp publication, man pages) a secure peer-to-peer high performance copy program supporting large windows, multiple streams, and bbftp a secure FTP program that supports large windows and multiple streams. These programs will be called from scripts to automate running multiple copies, gathering performance statistics, and reporting in real time and recording the results. We will also have scripts to gather SNMP data from the booth router. The over 35 remote sites are connected by various networks, including Internet 2, ESnet, JAnet, GARR, and Renater. We have identified and contacted sites and identified hosts that are suitable for the demonstration. The requirements for the hosts at the remote sites are fairly limited.
The information gathered will be recorded in files, and analyzed with various tools including Excel. Universal Time History (UTH) utility real time plots of the throughput and the aggregate throughput (see mock-up) to each remote site will be displayed on one monitor in the SLAC/FNAL booth. We will also record the routes and ping Round Trip Times (RTTs), losses, and derived throughputs among other metrics. We will display animations of RTT, loss and derived throughput measured by ping and throughput measured by iperf measured from SC2002 (also see screen-shot), iGrid2002 and from SLAC. One monitor will display a Java applet to show ping RTTs to the world in real-time from your computer (see the mock-up in case you cannot load the applet onto your computer). The other monitor will show available bandwidth estimates using packet pair techniques.
Longer term analysis (non-real time) will be made to summarize and report on the demonstration and will be made publicly available via the web. The web URL for this project will be: http://www-iepm.slac.stanford.edu/monitoring/bulk/sc2002/.
For a second demo at SC2002 we will illustrate the effects of Qbone Scavenger Service (QBSS), i.e. marking our traffic and routing it with LOWER priority than the other traffic. To do this we will use 3 Linux hosts with GE interfaces connected via a switch to a 1GE interface to the SC2002 show floor network. The idea is that two of the hosts will be able to saturate the 1GE interface and will run QBSS traffic. The third host will be alternately on and off, it will run Best Effort (i.e. unmakrked traffic). When the 3rd host runs it ought to be able to drive 600-900 Mbits/s and since its traffic is higher priority than the other 2 hosts, the other 2 hosts should back off. We will display the switch port utilization in real-time to show the effects.
Offsite resources will be at the sites listed in the table below. Each site will have one or more Unix hosts running iperf and bbftp servers. We are measuring throughputs to many of the sites from SLAC. We also measure routes and ping Round Trip Times (RTTs), losses, and derived throughputs among other metrics.
We will be located in the SLAC/FNAL booth.
We only require IP based communications.
No specialized on-show floor equipment is needed for the demonstration.
The following are the contacts at the various remote sites.
Ayumu Kubota, APAN-JP <firstname.lastname@example.org> Created August 6, 2002; last update August 8, 2002.
Linda Winkler, ANL, US, <email@example.com> + William E. Allcock [firstname.lastname@example.org]
Dantong Yu, BNL, Long Island, US, <email@example.com>
Harvey Newman, Caltech, Pasadena, US, <firstname.lastname@example.org> + Julian J. Bunn [email@example.com] + Suresh Singh <firstname.lastname@example.org>
Olivier Martin, CERN, Geneva, CH, <email@example.com> + Sylvain Ravot [Sylvain.Ravot@cern.ch]
Robin Tasker, Daresbury Lab, Liverpool, UK, <R.Tasker@dl.ac.uk> + Kummer, P. S (Paul) [P.S.Kummer@dl.ac.uk]
Jim Leighton, ESnet, Berkeley, US, <JFLeighton@lbl.gov>
Ruth Pordes, FNAL, Chicago, US, <firstname.lastname@example.org> + Frank Nagy <email@example.com> + Phil DeMar <firstname.lastname@example.org>
Andy Germain, NASA/GSFC, US, <email@example.com> + George Uhl [firstname.lastname@example.org]
Jerome Bernier, IN2P3, Lyon, FR, <email@example.com> + Dominique Boutigny [firstname.lastname@example.org]
Fabrizio Coccetti, INFN, Milan, IT, <Fabrizio Coccetti [email@example.com]>
Emanuele Leonardi, INFN, Rome, IT, <Emanuele.Leonardi@roma1.infn.it>
Guy Almes, Internet 2, US, <firstname.lastname@example.org> + Matt Zekauskas <email@example.com> + Stanislav Shalunov <firstname.lastname@example.org> + Ben Teitelbaum <email@example.com>
Chip Watson, JLab, Newport News, US, <firstname.lastname@example.org> + Robert Lukens <email@example.com>
Yukio Karita, KEK, Tokyo, JP, <firstname.lastname@example.org>, Teiji Nakamura <email@example.com>
Wu-chun Feng, LANL, Los Alamos, US, <firstname.lastname@example.org>, Mike Fisk <email@example.com>
Bob Jacobsen, LBL, Berkeley, US, <Bob_Jacobsen@lbl.gov>, Shane Canon <Canon@nersc.gov>
Richard Hughes-Jones, Manchester University, UK, <firstname.lastname@example.org>
Anthony Anthony, NIKHEF, Netherlands, <email@example.com>
Tom Dunigan, ORNL, Oak Ridge, US, <firstname.lastname@example.org> + Bill Wing <email@example.com>
Richard Baraniuk, Rice University, <firstname.lastname@example.org>, Rolf Riedi [email@example.com]
Takashi Ichihara, RIKEN, Japan, [firstname.lastname@example.org]
John Gordon, Rutherford Lab, Oxford, UK, <J.C.Gordon@RL.AC.UK> + Adye, TJ (Tim) [T.J.Adye@RL.AC.UK]
Reagan Moore, SDSC, San Diego, UA, <moore@SDSC.EDU> + Kevin Walsh [kwalsh@SDSC.EDU] + Arcot Rajasekar <sekar@SDSC.EDU>
Warren Matthews, SLAC, Menlo Park, US <email@example.com> + Paola Grosso <firstname.lastname@example.org> + Gary Buhrmaster <email@example.com> + Connie Logg <firstname.lastname@example.org> + Andy Hanushevsky <email@example.com> + Jerrod Williams <firstname.lastname@example.org> + Steffen Luitz <email@example.com>
Warren Matthews, Stanford University, Palo Alto, US, Milt Mallory <firstname.lastname@example.org>
William Smith, Sun Micro Systems [William.Smith@sun.com], Rocky Snyder email@example.com
Andrew Daviel, TRIUMF, Vancouver, CA, <firstname.lastname@example.org>
Yee-Ting Li, University College London, UK, <email@example.com> + Peter Clarke <firstname.lastname@example.org>
Constantinos Dovrolis, University of Delaware, US, <email@example.com>
Paul Avery, University of Florida, Gainesville, US, <firstname.lastname@example.org> + Gregory Goddard [email@example.com]
Thomas Hacker, University of Muchigan, <firstname.lastname@example.org>
Joe Izen, University of Texas at Dallas, US, <email@example.com>
Miron Livny, University of Wisconsin, Madison, US, <firstname.lastname@example.org> + Paul Barford <email@example.com> + Dave Plonka <firstname.lastname@example.org>>
Comments to email@example.com
Created August 6, 2002; last update August 8, 2002.