High Performance Mass Storage and Parallel I/O: Technologies and Applications

The definitive roadmap through the complex and fast–growing field of high performance I/O architecture
Today′s data–driven high performance computer technologies demand reliable delivery systems that combine high–level computing, storage, I/O, and network communication performance. Due to the growth of Internet–driven applications like digital libraries, virtual laboratories, video on demand, e–commerce, web services, and collaborative systems, issues such as storage capacity and access speed have become critical in the design of today′s computer systems.
High Performance Mass Storage and Parallel I/O fills the need for a readily accessible single reference source on the subject of high performance, large–scale storage and delivery systems, specifically the use of Redundant Arrays of Inexpensive Disks (RAID) that are accessed using parallel input/output (I/O) architecture. The authors, all internationally recognized experts in the field, have combined the best of the current literature on the subject with important information on emerging technologies and future trends.
Topics covered include:
∗ Redundant disk array Architecture
∗ Fault Tolerance Issues in Disk Arrays
∗ Caching and Prefetching
∗ Parallel File and I/O Systems
∗ Emerging Technologies and Future Trends
A valuable resource for both students of computer technology and professionals in the field, High Performance Mass Storage and Parallel I/O delivers state–of–the–art information that will help today′s system designers and application developers meet the increasing demand for high–performance, large–scale storage systems.

Spis treści:
Foreword.
Preface.
I. Introduction to Redundant Disk Array Architecture.
1. A Case for Redundant Arrays of Inexpensive Disks (RAID).
2. Disk System Architectures for High Performance Computing.< br>3. The Performance of Parity Placements in Disk Arrays.
4. A Performance Comparison of RAID–5 and Log–Structured Arrays.
II. Advanced Disk Array Architectures.
5. Parity Logging Overcoming the Small Write Problem in Redundant Disk Arrays.
6. Distributed RAID – A New Multiple Copy Algorithm.
7. The HP AutoRAID Hierarchical Storage System.
8. Scalable Distributed Log Structured Arrays.
9. Comparison of Sparing Alternatives for Disk Arrays.
10. Destage Algorithms for Disk Arrays with Non–Volatile Caches.
III. Fault Tolerance Issues in Disk Arrays.
11. Failure Correction Techniques for Large Disk Arrays.
12. Tolerating Multiple Failures in RAID Architectures with Optimal Storage and Uniform Declustering.
13. Parity Declustering for Continuous Operation in Redundant Disk Arrays.
14. The EVENODD Code and its Generalization.
IV. Caching and Prefetching.
15. RAPID–Cache – A Reliable and Inexpensive Write Cache for Disk I/O Systems.
16. Informed Prefetching and Caching.
17. Practical Prefetching Techniques for Multiprocessor File Systems.
18. Design Issues of a Cooperative Cache with no Coherence Problems.
19. Collective Buffering: Improving Parallel I/O Performance.
V. Parallel File Systems.
20. The Vesta Parallel File System.
21. The Zebra Striped Network File System.
22. PPFS: A High Performance Portable Parallel File System.
23. The Global File System.
24. Serverless Network File Systems.
VI. Parallel I/O Systems.
25. Parallel I/O Subsystems in Massively Parallel Supercomputer.
26. RAID–II: A High–Bandwidth Network File Server.
27. Petal: Distributed Virtual Disks.
28. A Cost–Effective, High–Bandwidth Storage Architecture.
29. RAID–x: A New Distributed Disk Array for I/O–Centric Cluster Computing.
30. Designing a Self–Maintaining Storage System.
31. Modeling and Eva luation of Fibre Channel Storage Area Networks.
VII. Parallel I/O Programming Paradigms.
32. Overview of the MPI–IO Parallel I/O Interface.
33. Disk Resident Arrays: An Array–Oriented I/O Library for Out–of–Core Computations.
34. Active Disks: Programming Model, Algorithms and Evaluation.
35. Disk–directed I/O for MIMD Multiprocessors.
VIII. Parallel I/O Applications and Environments.
36. Applications–Driven Parallel I/O.
37. Comparing Multimedia Storage Architectures.
38. High Availability in Clustered Multimedia Servers.
39. An Architecture for a Scalable High–Performance Digital Library.
40. I/O Requirements of Scientific Applications: An Evolutionary View.
41. Mitra: A Scalable Continuous Media Server.
IX. Emerging Technologies and Future Trends.
42. An Introduction to the InfiniBand Architecture.
43. XML, Hyper–media, and Fortran I/O.
44. I/O Programming Paradigms: Past and Future.
45. Scientific Applications using Parallel I/O.
Index.

Nota biograficzna:
HAI JIN is a professor of computer science at Huazhong University of Science and Technology, Wuhan, China. He holds both a B.A. and M.S. degree in computer science, and a Ph.D. in electrical and electronics engineering from the same University. He is currently a postdoctoral fellow in the Department of Electrical and Electronics Engineering at the University of Hong Kong in addition to being a visiting scholar at the Internet and Cluster Computing Laboratory at the University of Southern California. Dr. Jin has coauthored three books and published more than 50 papers in international journals and conferences.
TONI CORTES is an associate professor at Universitat Politecnica de Catalunya, Barcelona, Spain. He obtained his M.S. and Ph.D. degrees in computer science at the same university, and is currently the coordinator of the single–system image technical area in t he IEEE Task Force on Cluster Computing (TFCC). Dr. Cortes has also been working on several European industrial projects and has published more than 15 papers in international journals and conferences.
RAJKUMAR BUYYA is Co–Chair of the IEEE Task Force on Cluster Computing and an international speaker in the IEEE Computer Society Chapter Tutorials Program. Currently at Monash University, Melbourne, Australia, he is conducting R&D on the use of an economics paradigm for peer–to–peer and grid–based service–oriented computing. He has co–authored Microprocessor x86 Programming and Mastering C++, and edited a popular two–volume book on high performance cluster computing. He has published over 50 research articles in major international journals and conferences.

Okładka tylna:
The definitive roadmap through the complex and fast–growing field of high performance I/O architecture
Today′s data–driven high performance computer technologies demand reliable delivery systems that combine high–level computing, storage, I/O, and network communication performance. Due to the growth of Internet–driven applications like digital libraries, virtual laboratories, video on demand, e–commerce, web services, and collaborative systems, issues such as storage capacity and access speed have become critical in the design of today′s computer systems.
High Performance Mass Storage and Parallel I/O fills the need for a readily accessible single reference source on the subject of high performance, large–scale storage and delivery systems, specifically the use of Redundant Arrays of Inexpensive Disks (RAID) that are accessed using parallel input/output (I/O) architecture. The authors, all internationally recognized experts in the field, have combined the best of the current literature on the subject with important information on emerging technologies and future trends.
Top