The Green Computing Book -

. .

Editeur : Taylor & Francis Ltd

Langue : Anglais

Parution : 31/08/2020

Format : Moyen, de 350g à 1kg

Nombre de pages : 353

ISBN : 0367659158

Résumé :

Low-Power, Massively Parallel, Energy-Efficient Supercomputers. Compiler-Driven Energy Efficiency. An Adaptive Run-Time System for Improving Energy Efficiency. Energy-Efficient Multithreading through Run-Time Adaptation. Exploring Trade-Offs between Energy Savings and Reliability in Storage Systems. Cross-Layer Power Management. Energy-Efficient Virtualized Systems. Demand Response for Computing Centers. Implications of Recent Trends in Performance, Costs, and Energy Use for Servers.

...

Biographie:

Wu-chun Feng is a professor in the Department of Computer Science and Department of Electrical & Computer Engineering at Virginia Polytechnic Institute and State University, where he is the director of the Synergy Laboratory. He is also the principal investigator of mpiBlast and Supercomputing in Small Spaces, founder of MyVICE, and co-founder of the Green500 list. His research interests lie broadly at the synergistic intersection of computer architecture, systems software and middleware, and applications software.

Sommaire:

State-of-the-Art Approaches to Advance the Large-Scale Green Computing Movement

Edited by one of the founders and lead investigator of the Green500 list, The Green Computing Book: Tackling Energy Efficiency at Large Scale explores seminal research in large-scale green computing. It begins with low-level, hardware-based approaches and then traverses up the software stack with increasingly higher-level, software-based approaches.

In the first chapter, the IBM Blue Gene team illustrates how to improve the energy efficiency of a supercomputer by an order of magnitude without any system performance loss in parallelizable applications. The next few chapters explain how to enhance the energy efficiency of a large-scale computing system via compiler-directed energy optimizations, an adaptive run-time system, and a general prediction performance framework. The book then explores the interactions between energy management and reliability and describes storage system organization that maximizes energy efficiency and reliability. It also addresses the need for coordinated power control across different layers and covers demand response policies in computing centers. The final chapter assesses the impact of servers on data center costs.