Energy Efficient High Performance Processors: Recent Approaches for Designing Green High Performance Computing

ISBN-13: 9789811085536 / Angielski / Twarda / 2018 / 165 str.

Jawad Haj-Yahya; Avi Mendelson; Yosi Ben Asher

Opis

Widoczna okładka, to zdjęcie poglądowe, a rzeczywista szata graficzna może różnić się od prezentowanej. Nie gwarantujemy zgodności okładki z prezentowanym zdjęciem.

Spis treści

Introduction.- Background.- DOEE: Dynamic Optimization framework for better Energy Efficiency.- Fine-grain Power Breakdown of Modern Out-Of-Order Cores and its implications on Skylake based systems.- Compiler-Directed Power Management for Superscalars.- SEEM: Symbolic Execution for Energy Modeling.- Related Works.- Conclusions and Future Work.

O autorze

Jawad Haj-Yahya received his BSc degree in Computer Science from Technion – Israel Institute of Technology, and his MSc and PhD degrees in Computer Science from the University of Haifa. Jawad worked as a power management architect for high-performance processors (Sandy Bridge, Haswell, Skylake, etc) in the Processors’ Architecture group at the Intel Corporation for 13 years. Jawad’s honors include the Intel Achievement Award (IAA), which is the highest award at Intel. His research interests include energy aware computing, power estimation and applications, and low power IC design. Recently Jawad has joined Nanyang Technology University at Singapore as a cyber security research scientist.

Prof. Avi Mendelson has a blend of industrial and academic experience in several different areas, such as computer architecture, operating systems, power management, reliability, high-performance computing and hardware security. He received his PhD from the ECE Department, University of Massachusetts at Amherst (UMASS) in 1990 and his BSc and MSc degrees from the Computer Science Department, Technion. He was the manager of the academic outreach program at Microsoft R&D Israel, where he initiated various innovation-based activities for students. Before that, he worked for 11 years as a senior researcher and principal engineer at Intel. Among his achievements at Intel, he was the chief architect of the CMP (multicore-on-chip) feature of the first dual-core processors Intel developed, for which he received the Intel Achievement Award (the highest award at Intel). Mendelson has published more than 130 papers in refereed journals, and at conferences and workshops. He completed a full term as an associate editor of IEEE Computer Architecture Letters (CAL) and now serves as an associate editor of IEEE Transactions on Computers.

He served as program chair of a number of major conferences and as the general chair of the ISCA (International Symposium on Computer Architecture) in 2013. Recently he was elected to the Board of Governors of the IEEE Computer Society.

Prof. Yosi Ben-Asher received his PhD degree in Computer Science from the Hebrew University of Jerusalem. He is a professor at Computer Science Department, University of Haifa. He is currently leading several research projects including: P2NC a system to study probabilistic evaluation of Boolean circuits, a high-level synthesis compiler from C to Verilog with optimized memory layout, a silicon compiler that minimizes wire lengths, 1 K multicore chip and an ASIP compiler/CPU based on graph tiling. His research areas include compilers, EDA, parallel programming, ad-hoc networks and reconfigurable networks.

Prof. Anupam Chattopadhyay received his B.E. degree from Jadavpur University, India in 2000. He received his MSc from ALaRI, Switzerland and PhD from RWTH Aachen, Germany in 2002 and 2008 respectively. From 2008 to 2009, he worked as a member of consulting staff at CoWare R&D, Noida, India. From 2010 to 2014, he led the MPSoC Architectures Research Group at RWTH Aachen, as a junior professor. Since September 2014, he has been an Assistant Professor at SCE, NTU.

During his PhD, he worked on automatic RTL generation from the architecture description language LISA, which was later commercialized by a leading EDA vendor. He developed several high-level optimizations and verification flows for embedded processors. In his doctoral thesis, he proposed a language-based modeling, exploration and implementation framework for partially re-configurable processors. Together with his doctoral students, Anupam proposed domain-specific, high-level synthesis for cryptography, high-level reliability estimation flows for embedded processors, generalisation of classic linear algebra kernels and a novel multi-layered coarse-grained reconfigurable architecture. In these areas, he has published as a (co)-author over 100 conference/ journal papers, several book chapters and a book. Anupam has served on various TPCs of top conferences, regularly reviews journal/conference articles and presented multiple invited seminars/tutorials at prestigious venues. He is a member of ACM and a senior member of IEEE.

Anupam received Borcher's plaque from RWTH Aachen for his outstanding doctoral dissertation in 2008 and nomination for best IP award in DATE 2016.

Blurb

This book explores energy efficiency techniques for high-performance computing (HPC) systems using power-management methods. Adopting a step-by-step approach, it describes power-management flows, algorithms and mechanism that are employed in modern processors such as Intel Sandy Bridge, Haswell, Skylake and other architectures (e.g. ARM). Further, it includes practical examples and recent studies demonstrating how modem processors dynamically manage wide power ranges, from a few milliwatts in the lowest idle power state, to tens of watts in turbo state. Moreover, the book explains how thermal and power deliveries are managed in the context this huge power range.

 

The book also discusses the different metrics for energy efficiency, presents several methods and applications of the power and energy estimation, and shows how by using innovative power estimation methods and new algorithms modern processors are able to optimize metrics such as power, energy, and performance. Different power estimation tools are presented, including tools that break down the power consumption of modern processors at sub-processor core/thread granularity. The book also investigates software, firmware and hardware coordination methods of reducing power consumption, for example a compiler-assisted power management method to overcome power excursions. Lastly, it examines firmware algorithms for dynamic cache resizing and dynamic voltage and frequency scaling (DVFS) for memory sub-systems.