Description
Book SynopsisThis new guide covers everything you need to plan, build, and deploy a high-performance Linux cluster. You'll learn about planning, hardware choices, bulk installation of Linux on multiple systems, and other basic considerations. Learn about the major free software projects and how to choose those that are most helpful.
Table of ContentsPreface Part I. An Introduction to Clusters 1. Cluster Architecture Modern Computing and the Role of Clusters Types of Clusters Distributed Computing and Clusters Limitations; My Biases 2. Cluster Planning Design Steps Determining Your Cluster's Mission Architecture and Cluster Software Cluster Kits CD-ROM-Based Clusters Benchmarks 3. Cluster Hardware; Design Decisions Environment 4. Linux for Clusters Installing Linux Configuring Services Cluster Security Part II. Getting Started Quickly 5. openMosix What is openMosix? How openMosix Works Selecting an Installation Approach Installing a Precompiled Kernel Using openMosix Recompiling the Kernel; Is openMosix Right for You? 6. OSCAR Why OSCAR? What's in OSCAR Installing OSCAR Security and OSCAR Using switcher; Using LAM/MPI with OSCAR 7. Rocks Installing Rocks Managing Rocks Using MPICH with Rocks Part III. Building Custom Clusters; 8. Cloning Systems Configuring Systems Automating Installations; Notes for OSCAR and Rocks Users 9. Programming Software Programming Languages Selecting a Library LAM/MPI MPICH Other Programming Software Notes for OSCAR Users Notes for Rocks Users 10. Management Software C3 Ganglia Notes for OSCAR and Rocks Users 11. Scheduling Software OpenPBS Notes for OSCAR and Rocks Users 12. Parallel Filesystems PVFS; Using PVFS Notes for OSCAR and Rocks Users Part IV. Cluster Programming 13. Getting Started with MPI MPI A Simple Problem; An MPI Solution I/O with MPI Broadcast Communications; 14. Additional MPI Features More on Point-to-Point Communication; More on Collective Communication Managing Communicators; Packaging Data 15. Designing Parallel Programs Overview; Problem Decomposition Mapping Tasks to Processors Other Considerations 16. Debugging Parallel Programs Debugging and Parallel Programs Avoiding Problems Programming Tools; Rereading Code Tracing with printf Symbolic Debuggers; Using gdb and ddd with MPI Notes for OSCAR and Rocks Users; 17. Profiling Parallel Programs Why Profile? Writing and Optimizing Code Timing Complete Programs Timing C Code Segments; Profilers MPE Customized MPE Logging Notes for OSCAR and Rocks Users Part V. Appendix References Books URLs; Index
