Operating system power minimization through run-time processor resource adaptation

Li, Tao; John, Lizy K.

doi:10.1016/j.micpro.2005.12.001

Cited by 4 publications

(3 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Redstone et al [21] and Nellans et al [19] have shown that there are important classes of applications, namely webservers, databases, and display intensive applications for which the OS can contribute more than half the total instructions executed. Nellans et al [19] and Li et al [15] show OS execution under-performs user applications by 3-5x on modern out-of-order processors and suggest that OS code can be run on less aggressively designed processors to improve energy efficiency.…”

Section: Impact Of Os On System Throughputmentioning

confidence: 99%

“…This migration of OS execution to an energy-efficient core results in overall lower energy for the system. Li et al [15] take a slightly different approach to OS execution than other proposals. They propose that rather than implementing a secondary processor, existing uni-processors should be augmented so that aggressive out-of-order features can be throttled in the microarchitecture.…”

Section: Hardware Support For Efficient Os Executionmentioning

confidence: 99%

“…It also greatly simplifies the task of the OS developer, while incurring only a minor hardware storage overhead of about 2 KB. A related alternative to off-loading is the dynamic adaptation of a single processor's resources when executing OS sequences [15,20]. While the dynamic adaptation schemes do not rely on off-loading, the decision making process on when to adapt resources is very similar to that of making off-loading decisions.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Improving Server Performance on Multi-cores via Selective Off-Loading of OS Functionality

Nellans

Sudan

Brunvand

et al. 2011

Computer Architecture

View full text Add to dashboard Cite

Abstract. Modern and future server-class processors will incorporate many cores. Some studies have suggested that it may be worthwhile to dedicate some of the many cores for specific tasks such as operating system execution. OS off-loading has two main benefits: improved performance due to better cache utilization and improved power efficiency due to smarter use of heterogeneous cores. However, OS off-loading is a complex process that involves balancing the overheads of off-loading against the potential benefit, which is unknown while making the offloading decision. In prior work, OS off-loading has been implemented by first profiling system call behavior and then manually instrumenting some OS routines (out of hundreds) to support off-loading. We propose a hardware-based mechanism to help automate the off-load decisionmaking process, and provide high quality dynamic decisions via performance feedback. Our mechanism dynamically estimates the off-load requirements of the application and relies on a run-length predictor for the upcoming OS system call invocation. The resulting hardware based off-loading policy yields a throughput improvement of up to 18% over a baseline without off-loading, 13% over a static software based policy, and 23% over a dynamic software based policy.

show abstract

Section: Impact Of Os On System Throughputmentioning

confidence: 99%

Section: Hardware Support For Efficient Os Executionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation