Share This Author
The Quadrics Network: High-Performance Clustering Technology
The Quadrics network extends the native operating system in processing nodes with a network operating system and specialized hardware support in the network interface and provides network fault tolerance.
A Power-Aware Run-Time System for High-Performance Computing
A power-aware algorithm is proposed that automatically and transparently adapts its voltage and frequency settings to achieve significant power reduction and energy savings with minimal impact on performance in commodity HPC systems.
Inter-block GPU communication via fast barrier synchronization
This work proposes two approaches for inter-block GPU communication via barrier synchronization: GPU lock-based synchronization andGPU lock-free synchronization and evaluates the efficacy of each approach via a micro-benchmark as well as three well-known algorithms — Fast Fourier Transform, dynamic programming, and bitonic sort.
The design, implementation, and evaluation of mpiBLAST
This work presents the software architecture of mpiBLAST, an open-source parallelization of BLAST that achieves superlinear speed-up by segmenting a BLAST database and then having each node in a computational cluster search a unique portion of the database.
FAST TCP: from theory to experiments
A variant of TCP, called FAST, is described that can sustain high throughput and utilization at multigigabits per second over large distances and report the first experimental results.
CPU MISER: A Performance-Directed, Run-Time System for Power-Aware Clusters
- R. Ge, Xizhou Feng, Wu-chun Feng, K. Cameron
- Computer ScienceInternational Conference on Parallel Processing…
- 10 September 2007
Performance and power are critical design constraints in today's high-end computing systems. Reducing power consumption without impacting system performance is a challenge for the HPC community. We…
MOON: MapReduce On Opportunistic eNvironments
- Heshan Lin, Xiaosong Ma, J. Archuleta, Wu-chun Feng, M. Gardner, Zhe Zhang
- Computer ScienceHPDC '10
- 21 June 2010
Tests on an emulated volunteer computing system, which uses a 60-node cluster where each node possesses a similar hardware configuration to a typical computer in a student lab, demonstrate that MOON can deliver a three-fold performance improvement to Hadoop in volatile, volunteer computing environments.
The Green500 List: Encouraging Sustainable Supercomputing
The Green500 List effort will encourage the HPC community and operators of Internet data centers to design more power-efficient supercomputers and large-scale data centers and to supplement the TOP500 List.
On the energy efficiency of graphics processing units for scientific computing
- Song Huang, S. Xiao, Wu-chun Feng
- Computer ScienceIEEE International Symposium on Parallel…
- 23 May 2009
This paper seeks to characterize the notion of a “power-hungry GPU” via an empirical study of the performance, power, and energy characteristics of GPUs for scientific computing, and takes an important biological code that runs in a traditional CPU environment and transform and map it to a hybrid CPU+GPU environment.
Making a case for a Green500 list
- Sushant Sharma, C. Hsu, Wu-chun Feng
- Computer ScienceProceedings 20th IEEE International Parallel…
- 25 April 2006
It is argued that the TOP500 List makes it much more difficult for the high-performance computing (UPC) community to focus on performance metrics other than speed, and proposes a Green500 List to raise awareness to other performance metrics of interest.