Choose timezone
Your profile timezone:
CERN Computing Colloquium
Markets for resource allocation in distributed systems
by
→
Europe/Zurich
503/1-001 - Council Chamber (CERN)
Description
Peer to Peer clusters like the Grid and PlanetLab enable in principle
the same statistical multiplexing efficiency gains for computing as the
Internet provides for networking. The key unsolved problem is resource
allocation when user preferences vary over time and information about
the state of the system is not always perfect. Given that existing
solutions are not economically efficient and require high latency to
acquire resources, we designed and implemented Tycoon, a market based
distributed resource allocation system based on an Auction Share
scheduling algorithm. Preliminary results show that Tycoon achieves low
latency and high fairness while providing incentives for truth-telling
on the part of strategic users.
the same statistical multiplexing efficiency gains for computing as the
Internet provides for networking. The key unsolved problem is resource
allocation when user preferences vary over time and information about
the state of the system is not always perfect. Given that existing
solutions are not economically efficient and require high latency to
acquire resources, we designed and implemented Tycoon, a market based
distributed resource allocation system based on an Auction Share
scheduling algorithm. Preliminary results show that Tycoon achieves low
latency and high fairness while providing incentives for truth-telling
on the part of strategic users.
Bio:
Bernardo Huberman is a Senior HP Fellow and Director of the Information
Dynamics Lab at Hewlett Packard Laboratories. He received his Ph.D. in
Physics from the University of Pennsylvania, and is currently a
Consulting Professor in the Department of Applied Physics at Stanford
University. He has worked in condensed matter physics, dealing with
systems ranging from superionic conductors to two-dimensional
superfluids, and has made contributions to the theory of critical
phenomena in low dimensional systems. He is one of the discoverers of
chaos in a number of physical systems, and also established a number of
universal properties in nonlinear dynamical systems. His research into
the dynamics of complex structures led to his discovery of
ultradiffusion in hierarchical systems.
Dr. Huberman is a Fellow of the American Physical Society, a former
trustee of the Aspen Center for Physics and Fellow of the Japan Society
for the Promotion of Science, as well as a faculty member in the
Symbolic Systems Program at Stanford University. He is co-winner of the
1990 CECOIA prize in Economics and Artificial Intelligence and he
recently shared the IBM Prize of the Society for Computational
Economics. He was also the Chairman of the Council of Fellows at Xerox
Corporation and the manager of the Internet Ecologies Group. He has held
visiting professorships at the University of Paris, the University of
Copenhagen and the European School of Business.
Organiser(s): Jamie Shiers
More information: http://computing-colloquia.web.cern.ch/computing-colloquia/bios/huberman.htm