|« Previous Article||Next Article »|
Friday, May 21, 2010
By Scott Brandt
The ACM International Symposium on High Performance Distributed Computing will present its 2010 Best Paper award to Doctoral Candidate Povzner, Professor of Computer Science Brandt, and collaborator Darren Sawyer for the work "Horizon: Efficient Deadline-Driven Disk I/O Management for Distributed Storage Systems" at the annual conference on June 23.
Data centers often consolidate a variety of workloads to increase storage utilization and reduce management costs. Each workload has its own performance targets that need to be met, requiring isolation from the effects of other competing workloads sharing the system. Satisfying the global throughput and latency targets of each workload is challenging in fully distributed storage systems because workloads can have different data layouts and different requests from the same workload can be serviced by different nodes. New research in the UCSC Systems Research Laboratory (SRL) shows that a principled multi-layered approach can deliver global throughput and latency targets while efficiently utilizing system resources.
A paper on this research - "Horizon: Efficient Deadline-Driven Disk I/O Management for Distributed Storage Systems" - by CS Ph.D. student Anna Povzner, her advisor, CS Professor Scott Brandt, and their collaborator Darren Sawyer at NetApp was just selected to receive the Best Paper Award at the ACM International Symposium on High Performance Distributed Computing (HPDC 2010), the premier venue for presenting the latest research on the design, implementation, evaluation, and use of parallel and distributed systems for high performance and high end computing.
For more information about this research or the SRL, please contact Prof. Brandt (firstname.lastname@example.org). More information about HPDC 2010 can be found here: http://hpdc2010.eecs.northwestern.edu/. The program for the conference, including the best paper session, is available here: http://hpdc2010.eecs.northwestern.edu/program.html.