In Search of Clusters, 2/E
Gregory Pfister, IBM Corporation, Austin, Texas

ISBN-10: 0138997098
ISBN-13: 9780138997090

Publisher: Prentice Hall
Copyright: 1998
Format: Paper; 608 pp
Published: 12/12/1997

Suggested retail price: $44.95
Buy from myPearsonStore

This book has been revised and updated to cover the latest architectures, including new chapters on Availability, non-uniform memory access, and a completely revised chapter of examples.

The must-read guide to clustering technologies - for all network planners, software engineers, hardware designers, and anyone who purchases servers.

  • Completely updated! Extensive new coverage of availability and non-uniform memory access (NUMA).
  • Demonstrates the limitations of commercial server benchmarks.
  • Extensive coverage of cluster-related programming and architectural issues.

I. WHAT ARE CLUSTERS, AND WHY USE THEM?

1. Introduction.

Working Harder. Working Smarter. Getting Help. The Road to Lowly Parallel Processing. A Neglected Paradigm. What is to Come.

2. Examples.

Beer & Subpoenas. Serving the Web. The Farm. Fermilab. Other Compute Clusters. Full System Clusters. Cluster Software Products. Basic (Availability) Clusters. Not the End.

3. Why Clusters?

The Standard Litany. Why Now? Why Not Now? Commercial Node Performance. The Need for High Availability.

4. Definition, Distinctions, and Initial Comparisons.

Definition. Distinction from Parallel Systems. Distinctions from Distributed Systems. Concerning "Single System Image." Other Comparisons. Reactions.

II. HARDWARE.

5. A Cluster Bestiary.

Exposed vs. Enclosed. "Glass-House" vs. "Campus-Wide" Cluster. Cluster Hardware Structures. Communication Requirements. Cluster Acceleration Techniques.

6. Symmetric Multiprocessors.

What is an SMP? What is a Cache, and Why Is It Necessary? Memory Contention. Cache Coherence. Sequential and Other Consistencies. Input/Output. Summary.

7. NUMA and Friends.

UMA, NUMA, NORMA, and CC-NUMA. How CC-NUMA Works. The "N" in CC-NUMA. Software Implications. Other CC-NUMA Implications. Is "NUMA" Inevitable? Great Big CC-NUMA. Simple COMA.

III. SOFTWARE.

8. Workloads.

Why Discuss Workloads? Serial: Throughput. Parallel. Amdahl's Law. The Point of All This.

9. Basic Programming Models and Issues.

What is a Programming Model? The Sample Problem. Uniprocessor. Shared Memory. Message-Passing. CC-NUMA. SIMD and All That. Importance.

10. Commercial Programming Models.

Small N vs. Large N. Small N Programming Models. Large-N I/O Programming Models. Large-N Processor-Memory Models. Shared Disk or not Shared Disk?

11. Single System Image.

Single System Image Boundaries. Single System Image Levels. The Application and Subsystem Levels. The Operating System Kernel Levels. Hardware Levels. SSI and System Management.

IV. SYSTEMS.

12. High Availability.

What Does "High Availability" Mean? The Basic Idea: Failover. Resources. Failing Over Data. Failing Over Communications. Towards Instant Failover. Failover to Where? Lock Data Reconstruction. Heartbeats, Events, and Failover Processing. System Structure. Related Issues.

13. Symmetric Multiprocessors, "NUMA," and Clusters.

Preliminaries. Performance. Cost. High Availability. Other Issues. Partitioning. Conclusion.

14. Why We Need the Concept of Cluster.

Benchmarks. Development Directions. Confusion of Issues. The Lure of Large Numbers.

15. Conclusion.

Cluster Operating Systems. Exploitation. Standards. Software Pricing. What About 2010?. Coda: The End of Parallel Computer Architecture.

Annotated Bibliography.

Index.

About the Author.

Simon Guerrero (w-beard@netcomuk.co.uk) from Stone, Staffordshire UK , 04/23/98, rating=10:

Learn about clusters without falling asleep

About a month ago I started work on a project running on a small cluster and involving the Oracle 8 Parallel Server at a low level (writing the Distributed Lock Manager support libraries for a certain OS). At this point, I'd never used (or even seen!) a clustered system, and I knew nothing about clusters at all. Then a colleague loaned me the first edition of Dr Pfister's book. Unwilling to be over-eager to learn anything out of 'paid' time, I opened the book with some trepidation, expecting to find the usual dessicated prose and tons of TLAs. What a pleasant surprise! From the 'legal stuff' at the front of the book ('a kind of garlic'), right through to the bibliography ('I found this paper almost unreadable'), the author understands the need of the reader to remain conscious through what is potentially the dullest of subjects and emerge, slightly surprised ('Did I actually enjoy that?') at the other end. Thousands of college lecturers have a lot to learn from this man!

The second edition of the book is more a re-write than an update, and just as packed with anecdotes, humour (right down to pseudo-Paul Simon lyrics - people were hanged for less in the Wild West), and at the same time, probably the most thorough explanations of the why/how/when/wheres of clustering you will find in any book. As the quote on the back says 'This book is what would happen if Scott Adams wrote a book on parallel computers'... Full marks!

GREGORY PFISTER is an IBM Senior Technical Staff Member living in Austin, TX, and an internationally-recognized expert on parallel processing and high availability. His S.B., S.M., and Ph.D. were earned at MIT. He has been an instructor at MIT, an assistant professor at U.C. Berkeley, and a member of the R&D staff of IBM for many years. Two of his many published papers have won "best paper" awards at major conferences.


89970-8

The #1 guide to clustering for every IT professional!

From Microsoft to IBM, Compaq to Sun to DEC, virtually every large computer company now uses clustering as a key strategy for high-availability, high-performance computing. This book tells you why-and how. It cuts through the marketing hype and techno-religious wars surrounding parallel processing, delivering the practical information you need to purchase, market, plan or design servers and other high-performance computing systems.

  • Microsoft Cluster Services ("Wolfpack")
  • IBM Parallel Sysplex and SP systems
  • DEC OpenVMS Cluster and Memory Channel
  • Tandem ServerNet and Himalaya
  • Intel Virtual Interface Architecture
  • Symmetric Multiprocessors (SMPs) and NUMA systems

With unequalled simplicity, directness, and humor, expert Gregory Pfister delivers all the information you need to make critical strategic decisions. He introduces the primary technologies involved in clustering, and shows why they are becoming so important. He compares clustering with symmetric multiprocessing, demonstrating major differences that are often "papered over." The legendary first edition of this book predicted the cluster revolution now underway. Its refreshing style, candid opinions, and simple explanations made it an underground classic. This new edition adds more than 150 pages of new material, including detailed new coverage of high availability, and Non-Uniform Memory Access (NUMA).

View a Sample Chapter PDF:

Pearson Higher Education offers special pricing when you choose to package your text with other student resources. If you're interested in creating a cost-saving package for your students, contact your Pearson Higher Education representative for pricing and ordering information.

Pearson Higher Education offers special pricing when you choose to package your text with other student resources. If you're interested in creating a cost-saving package for your students contact your Pearson Higher Education representative.


Copyright ©2008 Pearson Education. All rights reserved. Legal Notice | Privacy Policy | Permissions