Sciweavers

828 search results - page 125 / 166
» Modeling of Workload in MPPs
Sort
View
SC
2009
ACM
15 years 6 months ago
Evaluating the impact of inaccurate information in utility-based scheduling
Proponents of utility-based scheduling policies have shown the potential for a 100–1400% increase in value-delivered to users when used in lieu of traditional approaches such as...
Alvin AuYoung, Amin Vahdat, Alex C. Snoeren
DATE
2009
IEEE
155views Hardware» more  DATE 2009»
15 years 6 months ago
Using non-volatile memory to save energy in servers
Abstract—Recent breakthroughs in circuit and process technology have enabled new usage models for non-volatile memory technologies such as Flash and phase change RAM (PCRAM) in t...
David Roberts, Taeho Kgil, Trevor N. Mudge
ICC
2009
IEEE
154views Communications» more  ICC 2009»
15 years 6 months ago
Load Balancing vs. Distributed Rate Limiting: An Unifying Framework for Cloud Control
— With the expansion of cloud-based services, the question as to how to control usage of such large distributed systems has become increasingly important. Load balancing (LB), an...
Rade Stanojevic, Robert Shorten
ICCS
2009
Springer
15 years 6 months ago
A Holistic Approach for Performance Measurement and Analysis for Petascale Applications
Abstract. Contemporary high-end Terascale and Petascale systems are composed of hundreds of thousands of commodity multi-core processors interconnected with high-speed custom netwo...
Heike Jagode, Jack Dongarra, Sadaf R. Alam, Jeffre...
SIGCOMM
2009
ACM
15 years 6 months ago
Understanding TCP incast throughput collapse in datacenter networks
TCP Throughput Collapse, also known as Incast, is a pathological behavior of TCP that results in gross under-utilization of link capacity in certain many-to-one communication patt...
Yanpei Chen, Rean Griffith, Junda Liu, Randy H. Ka...