Panel Discussion #1
Wednesday 6:30 - 8:00 pm

Impact of Grid Computing on Network Operators and HW Vendors

Grid Computing is an attempt to make computing work like the power grid.   When you run a job, you shouldn't know or care where it runs, so long as it gets done within your constraints (including security).   However, in attempting to accomplish this, Grid researchers are presenting network access patterns and loads different from what has been typical of Internet traffic.   MPI applications are looking for latency critical, bursty, small message traffic, some applications are producing data sets in the 100's of GBs and even Terabytes that need to be moved quickly and efficiently, or you might need remote control of earthquake shake tables and thus require constant jitter.   Grid researchers are asking for finer grained control of the network, dynamic optical routes,   allowing user apps (via middleware) to alter router configurations, etc..   For some network operators, this sounds like their worst nightmare come true.   For the network HW vendors, this presents challenges to say the least.

This panel is intended to bring together Grid researchers, network operators, and network HW vendors to discuss what the Grid researchers want and why, what impact that will have on network operations, and what challenges it will bring for the future HW designs.

Panel moderator:  
Brian L. Tierney is a Staff Scientist and group leader in the Distributed Systems Department at Lawrence Berkeley National Laboratory. His research interests include monitoring and troubleshooting techniques for widely distributed systems (Grids), and high-speed networking issues. Mr. Tierney has an M.S. in Computer Science from San Francisco State University, and a B.A. in physics from the University of Iowa. http://dsd.lbl.gov/~tierney/

Panelists:

Bill St. Arnaud, Senior Director, Advanced Networks for CANARIE Inc., Canada's Advanced Internet Development Organization. http://www.canarie.ca/~bstarn/

Tal Lavian, Principal Scientist, Advanced Technology Research, CTO Office, Nortel Networks. Tal is active in Grid Computing and resource orchestration of network services and specifically in optical networks environment.   http://www.nortel.com/drac

Bill Johnston, ESnet Manager and Senior Scientist, Information Technologies and Services Division Lawrence Berkeley National Laboratory http://dsd.lbl.gov/~johnston/

Phil Papadopoulos, Program Director, Grid and Cluster Computing and Acting Group Leader, Grid Development and Deployment, San Diego Supercomputing Center. http://users.sdsc.edu/~phil/homepage.html

Masum Z. Hasan,   Senior Technical Leader, Cisco Systems, USA. Masum is working on software and networking aspects of Grid Networking. He has a Masters in Computer Engineering from Odessa Polytechnic University in former USSR, and MMath and PhD in Computer Science from University of Waterloo, Canada.

A panel will discuss how data centers and high performance computing systems currently use two main classes of interconnection networks. Those based on the Ethernet protocol (EtherNET) and those with proprietary protocols (EtherNOT means Infiniband, Infiniband variants such as PathScale and Cray XD1, Quadrics, Myricom, etc. ).

While Ethernet can be considered a commodity product, other specialized systems can achieve better performance and can scale to configurations of several thousands of nodes, for example, by providing special support for collective communication.