17th Workshop on Job Scheduling Strategies for Parallel Processing (JSSPP)
In Conjunction with
IPDPS 2013
Boston, Massachusetts, USA
24 May 2013
Workshop organizers
Walfredo Cirne, Google
Narayan Desai, Argonne National Laboratory
Program Committee
Henri Casanova, University of Hawaii at Manoa
Julita Corbalan, Technical University of Catalonia
Dick Epema, Delft University of Technology
Gilles Fedak, INRIA
Dror Feitelson, The Hebrew University
Liana Fong, IBM T. J. Watson Research Center
Eitan Frachtenberg, Facebook
Alfredo Goldman, University of Sao Paulo
Allan Gottlieb, New York University
Alexandru Iosup, Delft University of Technology
Morris Jette, SchedMD LLC
Rajkumar Kettimuthu, Argonne National Laboratory
Dalibor Klusáček, Masaryk University
Zhiling Lan, Illinois Institute of Technology
Bill Nitzberg, Altair Engineering
David Oppenheimer, Google
Uwe Schwiegelshohn, Technical University Dortmund
Mark Squillante, IBM T. J. Watson Research Center
Wei Tang, Argonne National Laboratory
Dan Tsafrir, Technion - Israel Institute of Technology
Ramin Yahyapour, GWDG - University of Göttingen
Keynote:
Opportunistic Computing in Dynamic Compute Markets Stephen Elliott (Senior Manager at Amazon Web Services)
Owners of compute clusters have
always been concerned about maximizing
their resource utilization. Cloud Computing's value proposition to
customers is that they no longer need to invest significant capital
expenditures up front to build data centers based on speculative
demand forecasts. This capacity planning problem has been shifted to
Cloud Computing providers who can take advantage of (1) economies of
scale and scope to build and operate more cost-effective data centers
and (2) “statistics of scale” to achieve smoother aggregate demand
curves and better resource utilization (e.g., by serving customers and
applications diversified across locations, time zones, and
applications). To continue to lower prices while offering better
computing services and performance to customers, providers must think
of new products and services to optimize resource utilization. They
have to think beyond the basic capacity planning question of "How much
do I think customers will request at peak?" and explore questions like
"What products can I introduce to make otherwise idle computing
capacity useful to customers?" This talk will describe products
introduced by Amazon Web Services to promote “Opportunistic
Computing”--for example, Spot Instances, on which an ad hoc
supercomputing cluster performed 39-core years of scientific computing
in 11 hours on spare compute capacity--and highlight areas for further
research and development.
Multi Resource Fairness: Problems and Challenges Dalibor Klusáček (Masaryk University, Czech Republic),
Hana Rudová (Masaryk University, Czech Republic),
and Michal Jaroš (Institute of Computer Science, Masaryk University, Czech Republic)
A Periodic Portfolio Scheduler for Scientific Computing in the Data Center Kefeng Deng (National University of Defense Technology, P.R. China),
Ruben Verboon (Delft University of Technology, The Netherlands),
Kaijun Ren (National University of Defense Technology, P.R. China),
and Alexandru Iosup (Delft University of Technology, The Netherlands)
The proceedings of this workshop were published by Springer-Verlag
in the Lecture Notes on Computer Science series, as volume 8429.
It is also available from Springer on-line, with the option to
purchase single papers.