Fine-grained power management using process-level profiling

Cai, Hui; Li, Youhuizi; Shi, Wei

doi:10.1016/j.suscom.2012.01.002

Cited by 22 publications

(12 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The main part of power consumed by a server is accounted for the CPU, followed by the memory [26]. Based on this, Chen and Shi [10] present a process-level power profiling tool and a power-aware system module that eliminates energy wasted by abnormal-behavior applications for which hardware information is essential. The authors encourage the design of simple energy models to obtain real and instant measurements to control energy consumed by applications.…”

Section: Related Workmentioning

confidence: 99%

Reducing energy usage in resource-intensive Java-based scientific applications via micro-benchmark based code refactorings

et al. 2019

View full text Add to dashboard Cite

In-silico research has grown considerably. Today's scientific code involves long-running computer simulations and hence powerful computing infrastructures are needed. Traditionally, research in high-performance computing has focused on executing code as fast as possible, while energy has been recently recognized as another goal to consider. Yet, energy-driven research has mostly focused on the hardware and middleware layers, but few efforts target the application level, where many energy-aware optimizations are possible. We revisit a catalog of Java primitives commonly used in OO scientific programming, or micro-benchmarks, to identify energy-friendly versions of the same primitive. We then apply the micro-benchmarks to classical scientific application kernels and machine learning algorithms for both single-thread and multi-thread implementations on a server. Energy usage reductions at the micro-benchmark level are substantial, while for applications obtained reductions range from 3.90% to 99.18%.

show abstract

Section: Related Workmentioning

confidence: 99%

Reducing energy usage in resource-intensive Java-based scientific applications via micro-benchmark based code refactorings

et al. 2019

View full text Add to dashboard Cite

show abstract

“…With the increasing number of everyday online services that rely on data centres, energy and cost efficiency for data centres has recently received a significant amount of attention in the research community [11], [12], [13]. Here we focus specifically on power-cycling techniques.…”

Section: Related Workmentioning

confidence: 99%

Real-Time Power Cycling in Video on Demand Data Centres Using Online Bayesian Prediction

Marco

Wang

Porter

2017

2017 IEEE 37th International Conference on Distributed Computing Systems (ICDCS)

View full text Add to dashboard Cite

Energy usage in data centres continues to be a major and growing concern as an increasing number of everyday services depend on these facilities. Research in this area has examined topics including power smoothing using batteries and deep learning to control cooling systems, in addition to optimisation techniques for the software running inside data centres. We present a novel real-time power-cycling architecture, supported by a media distribution approach and online prediction model, to automatically determine when servers are needed based on demand. We demonstrate with experimental evaluation that this approach can save up to 31% of server energy in a cluster. Our evaluation is conducted on typical rack mount servers in a data centre testbed and uses a recent real-world workload trace from the BBC iPlayer, an extremely popular video on demand service in the UK.

show abstract

“…The radio and Wi-Fi power is based on the signal states and packets rate respectively. More details of the specific power models can be found in [8]. In the average power models, all c are constants and u are the usage of each component.…”

Section: B Lifetime Prediction Modelmentioning

confidence: 99%

One charge for one week: Hype or reality?

Luo

Chen

et al. 2014

International Green Computing Conference

Self Cite

View full text Add to dashboard Cite

Abstract-For mobile devices, battery energy is the most precious resource. In the last decade, researchers have proposed various energy saving strategies from the system level to the hardware component level. In this paper, we explore how close to one week a smart phone running off of a single battery can last under normal usage. We first developed a battery lifetime prediction model that considers the influence of both user behavior and hardware components. Through experiments we analyzed the assumptions and the accuracy of the prediction model. We discussed the error rate of the estimated applications' power as well as its influence on the battery lifetime prediction. To analyze the impact generated by user behavior, we classify users into six types based on their application usage pattern. The theoretical battery life and potential extended battery time for each user type, with and without hardware improvement, have been illustrated. For example, compared with the original 66h (2.75 days) for users who rarely use their smartphone, we found that the battery life can be extended to 147h (more than 6 days) when we only maintain applications in the top three commonly used categories. Finally, several aspects, such as sleep frequency and background applications, that may affect prediction results are discussed.

show abstract

Fine-grained power management using process-level profiling

Cited by 22 publications

References 24 publications

Reducing energy usage in resource-intensive Java-based scientific applications via micro-benchmark based code refactorings

Reducing energy usage in resource-intensive Java-based scientific applications via micro-benchmark based code refactorings

Real-Time Power Cycling in Video on Demand Data Centres Using Online Bayesian Prediction

One charge for one week: Hype or reality?

Contact Info

Product

Resources

About