A survey on techniques for improving the energy efficiency of large-scale distributed systems

Orgerie, Anne‐Cécile; Assunção, Marcos Dias de; Lefèvre, Laurent

doi:10.1145/2532637

Cited by 262 publications

(173 citation statements)

References 118 publications

Supporting

Mentioning

171

Contrasting

Unclassified

Order By: Relevance

“…This is a central issue because, through current technologies, an exascale machine would consume more power than what is reasonable [46,67]. It should also be a concern for parallel I/O researchers as, although the processing units are responsible for most of the power demand in typical computational systems, many applications spend most of their time performing I/O operations [11,78]. For this reason, increasing the energy efficiency of the I/O subsystem is also an important step to tackle the energy and power challenge.…”

Section: Resultsmentioning

confidence: 99%

A Checkpoint of Research on Parallel I/O for High-Performance Computing

et al. 2018

View full text Add to dashboard Cite

We present a comprehensive survey on parallel I/O in the high performance computing (HPC) context. This is an important field for HPC because of the historic gap between processing power and storage latencies, which causes applications performance to be impaired when accessing or generating large amounts of data. As the available processing power and amount of data increase, I/O remains a central issue for the scientific community. In this survey, we focus on a traditional I/O stack, with a POSIX parallel file system. We present background concepts everyone could benefit from. Moreover, through the comprehensive study of publications from the most important conferences and journals in a five-year time window, we discuss the state of the art of I/O optimization approaches, access pattern extraction techniques, and performance modeling, in addition to general aspects of parallel I/O research. Through this approach, we aim at identifying the general characteristics of the field and the main current and future research topics.

show abstract

Section: Resultsmentioning

confidence: 99%

A Checkpoint of Research on Parallel I/O for High-Performance Computing

et al. 2018

View full text Add to dashboard Cite

show abstract

“…This is another different line that recently has renewed researchers' interest and that in the near future could offer some novelties. Compiling, scheduling, and load balancing algorithms have a profound impact on performances and power consumption, leaving programmers free to focus on their own problems instead of caring for code optimizations [140,141].…”

Section: Where We Are Going To: Some Trends In High-performance Compumentioning

confidence: 99%

High‐performance computing: to boldly go where no human has gone before

Limet

Smari

Spalazzi

2015

Concurrency and Computation

View full text Add to dashboard Cite

International audienceComputing and computational science are well on their way to enter into the exascale era while High Performance Computing (HPC) is now present in many spheres and domains of modern society. New technologies such as cloud computing, big data, and connected objects open many perspectives and possibilities that were unattainable until now. These are high times for computing and for HPC in particular. The High Performance Computing \& Simulation conference aims to bring together researchers working on various aspects of HPC, their design and use, and their applications. As has been the conference tradition, selected extended papers of HPCS 2012 are presented in this special issue. These should be interesting contributions that supplement the current state-of-the-art research in HPC. To relate these works and highlight their value, in this article, we briefly trace back the history of HPC, sketch the state of the present research and development in the field, and project some of the challenges and future trends anticipated

show abstract

“…In [18], authors present a survey on techniques for improving the energy efficiency of large-scale distributed systems. A taxonomy and survey of energy-efficient data centers and cloud computing systems can be found in [19].…”

Section: Related Workmentioning

confidence: 99%

MERCi-MIsS: Should I Turn off My Servers?

Callau-Zori

Arantes

Sopena

et al. 2015

Distributed Applications and Interoperable Systems

View full text Add to dashboard Cite

Abstract. In recent years, the electrical consumption of data centers has increased considerably leading to a rise in the expenditure bill and in greenhouse gas emissions. Several existing on/off algorithms reduce energy consumption in data centers or Clouds by turning off unused (idle) machines. However, the turning off/on of servers consumes a certain amount of energy and also induces the wear and tear of disks. Based on the data streaming paradigm which deals with large amount of data on-line, we present in this paper MERCi-MIsS, a proposal whose aim is to save energy in data centers and Clouds and tackle the above tradeoff problems without degrading, as much as possible, the quality of services of the system. MERCi-MIsS dynamically estimates the future workload based on the recent past workload, deciding if servers should then be turned either on or off. We have implemented MERCi-MIsS on top of Twitter Storm. Evaluation results from experiments using real traces from Grid'5000 confirm the effectiveness and efficiency of MERCi-MIsS algorithm to save energy and avoid disk damage while the quality of service is only slightly degraded.

show abstract

A survey on techniques for improving the energy efficiency of large-scale distributed systems

Cited by 262 publications

References 118 publications

A Checkpoint of Research on Parallel I/O for High-Performance Computing

A Checkpoint of Research on Parallel I/O for High-Performance Computing

High‐performance computing: to boldly go where no human has gone before

MERCi-MIsS: Should I Turn off My Servers?

Contact Info

Product

Resources

About