High Performance Linux Clusters with OSCAR, Rocks, OpenMosix, and MPI - Helion

ebook

Autor: Joseph D Sloan
ISBN: 978-05-965-5245-9
stron: 370, Format: ebook
Data wydania: 2004-11-16
Księgarnia: Helion

Cena książki: 118,15 zł (poprzednio: 137,38 zł)
Oszczędzasz: 14% (-19,23 zł)

Osoby, które kupiły tę książkę, wybierały także »

Tagi: Linux

To the outside world, a "supercomputer" appears to be a single system. In fact, it's a cluster of computers that share a local area network and have the ability to work together on a single problem as a team. Many businesses used to consider supercomputing beyond the reach of their budgets, but new Linux applications have made high-performance clusters more affordable than ever. These days, the promise of low-cost supercomputing is one of the main reasons many businesses choose Linux over other operating systems.This new guide covers everything a newcomer to clustering will need to plan, build, and deploy a high-performance Linux cluster. The book focuses on clustering for high-performance computation, although much of its information also applies to clustering for high-availability (failover and disaster recovery). The book discusses the key tools you'll need to get started, including good practices to use while exploring the tools and growing a system. You'll learn about planning, hardware choices, bulk installation of Linux on multiple systems, and other basic considerations. Then, you'll learn about software options that can save you hours--or even weeks--of deployment time.Since a wide variety of options exist in each area of clustering software, the author discusses the pros and cons of the major free software projects and chooses those that are most likely to be helpful to new cluster administrators and programmers. A few of the projects introduced in the book include:

MPI, the most popular programming library for clusters. This book offers simple but realistic introductory examples along with some pointers for advanced use.
OSCAR and Rocks, two comprehensive installation and administrative systems
openMosix (a convenient tool for distributing jobs), Linux kernel extensions that migrate processes transparently for load balancing
PVFS, one of the parallel filesystems that make clustering I/O easier
C3, a set of commands for administering multiple systems

Ganglia, OpenPBS, and cloning tools (Kickstart, SIS and G4U) are also covered. The book looks at cluster installation packages (OSCAR & Rocks) and then considers the core packages individually for greater depth or for folks wishing to do a custom installation. Guidelines for debugging, profiling, performance tuning, and managing jobs from multiple users round out this immensely useful book.

Osoby które kupowały "High Performance Linux Clusters with OSCAR, Rocks, OpenMosix, and MPI", wybierały także:

Administracja systemem Linux. Kurs video. Przewodnik dla początkujących 59,00 zł, (17,70 zł -70%)
Gray Hat C#. Język C# w kontroli i łamaniu zabezpieczeń 57,74 zł, (17,90 zł -69%)
Bezpiecze 129,00 zł, (64,50 zł -50%)
Python dla administrator 178,98 zł, (89,49 zł -50%)
Linux. Wiersz polece 149,00 zł, (74,50 zł -50%)

Spis treści

High Performance Linux Clusters with OSCAR, Rocks, OpenMosix, and MPI eBook -- spis treści

High Performance Linux Clusters with OSCAR, Rocks, OpenMosix, and MPI
- SPECIAL OFFER: Upgrade this ebook with OReilly
- A Note Regarding Supplemental Files
- Preface
  - Audience
  - Organization
  - Conventions
  - How to Contact Us
  - Using Code Examples
  - Acknowledgments
- I. An Introduction to Clusters
  - 1. Cluster Architecture
    - 1.1. Modern Computing and the Role of Clusters
      - 1.1.1. Uniprocessor Computers
      - 1.1.2. Multiple Processors
        
        1.1.2.1. Centralized multiprocessors
        
        1.1.2.2. Multicomputers
        
        1.1.2.3. Cluster structure
    - 1.2. Types of Clusters
    - 1.3. Distributed Computing and Clusters
    - 1.4. Limitations
      - 1.4.1. Amdahls Law
    - 1.5. My Biases
  - 2. Cluster Planning
    - 2.1. Design Steps
    - 2.2. Determining Your Cluster's Mission
      - 2.2.1. What Is Your User Base?
      - 2.2.2. How Heavily Will the Cluster Be Used?
      - 2.2.3. What Kinds of Software Will You Run on the Cluster?
      - 2.2.4. How Much Control Do You Need?
      - 2.2.5. Will This Be a Dedicated or Shared Cluster?
      - 2.2.6. What Resources Do You Have?
      - 2.2.7. How Will Cluster Access Be Managed?
      - 2.2.8. What Is the Extent of Your Cluster?
      - 2.2.9. What Security Concerns Do You Have?
    - 2.3. Architecture and Cluster Software
      - 2.3.1. System Software
      - 2.3.2. Programming Software
      - 2.3.3. Control and Management
    - 2.4. Cluster Kits
    - 2.5. CD-ROM-Based Clusters
      - 2.5.1. BCCD
    - 2.6. Benchmarks
  - 3. Cluster Hardware
    - 3.1. Design Decisions
      - 3.1.1. Node Hardware
        
        3.1.1.1. CPUs and motherboards
        
        3.1.1.2. Memory and disks
        
        3.1.1.3. Monitors, keyboards, and mice
        
        3.1.1.4. Adapters, power supplies, and cases
      - 3.1.2. Cluster Head and Servers
      - 3.1.3. Cluster Network
    - 3.2. Environment
      - 3.2.1. Cluster Layout
      - 3.2.2. Power and Air Conditioning
        
        3.2.2.1. Power
        
        3.2.2.2. HVAC
      - 3.2.3. Physical Security
  - 4. Linux for Clusters
    - 4.1. Installing Linux
      - 4.1.1. Selecting a Distribution
      - 4.1.2. Downloading Linux
      - 4.1.3. What to Install?
    - 4.2. Configuring Services
      - 4.2.1. DHCP
      - 4.2.2. NFS
        
        4.2.2.1. Running NFS
        
        4.2.2.2. Automount
      - 4.2.3. Other Cluster File System
      - 4.2.4. SSH
        
        4.2.4.1. Using SSH
      - 4.2.5. Other Services and Configuration Tasks
        
        4.2.5.1. Apache
        
        4.2.5.2. Network Time Protocol (NTP)
        
        4.2.5.3. Virtual Network Computing (VNC)
        
        4.2.5.4. Multicasting
        
        4.2.5.5. Hosts file and name services
    - 4.3. Cluster Security
- II. Getting Started Quickly
  - 5. openMosix
    - 5.1. What Is openMosix?
    - 5.2. How openMosix Works
    - 5.3. Selecting an Installation Approach
    - 5.4. Installing a Precompiled Kernel
      - 5.4.1. Downloading
      - 5.4.2. Installing
      - 5.4.3. Configuration Changes
    - 5.5. Using openMosix
      - 5.5.1. User Tools
        
        5.5.1.1. mps and mtop
        
        5.5.1.2. migrate
        
        5.5.1.3. mosctl
        
        5.5.1.4. mosmon
        
        5.5.1.5. mosrun
        
        5.5.1.6. setpe
      - 5.5.2. openMosixView
      - 5.5.3. Testing openMosix
    - 5.6. Recompiling the Kernel
    - 5.7. Is openMosix Right for You?
  - 6. OSCAR
    - 6.1. Why OSCAR?
    - 6.2. What's in OSCAR
    - 6.3. Installing OSCAR
      - 6.3.1. Prerequisites
      - 6.3.2. Network Configuration
      - 6.3.3. Loading Software on Your Server
      - 6.3.4. A Basic OSCAR Installation
        
        6.3.4.1. Step 0: Downloading additional packages
        
        6.3.4.2. Step 1: Package selection
        
        6.3.4.3. Step 2: Configuring packages
        
        6.3.4.4. Step 3: Installing server software
        
        6.3.4.5. Step 4: Building a client image
        
        6.3.4.6. Step 5: Defining clients
        
        6.3.4.7. Step 6: Setting up the network
        
        6.3.4.8. Step 7: Completing the setup
        
        6.3.4.9. Step 8: Testing
      - 6.3.5. Custom Installations
      - 6.3.6. Changes OSCAR Makes
      - 6.3.7. Making Changes
    - 6.4. Security and OSCAR
      - 6.4.1. pfilter
      - 6.4.2. SSH and OPIUM
    - 6.5. Using switcher
    - 6.6. Using LAM/MPI with OSCAR
  - 7. Rocks
    - 7.1. Installing Rocks
      - 7.1.1. Prerequisites
      - 7.1.2. Downloading Rocks
      - 7.1.3. Installing the Frontend
      - 7.1.4. Install Compute Nodes
      - 7.1.5. Customizing the Frontend
        
        7.1.5.1. User management with 411
        
        7.1.5.2. X Window System
      - 7.1.6. Customizing Compute Nodes
        
        7.1.6.1. Adding packages
        
        7.1.6.2. Changing disk partitions
        
        7.1.6.3. Other changes
    - 7.2. Managing Rocks
    - 7.3. Using MPICH with Rocks
- III. Building Custom Clusters
  - 8. Cloning Systems
    - 8.1. Configuring Systems
      - 8.1.1. Distributing Files
        
        8.1.1.1. Pushing files with rsync
    - 8.2. Automating Installations
      - 8.2.1. Kickstart
        
        8.2.1.1. Configuration file
        
        8.2.1.2. Using Kickstart
      - 8.2.2. g4u
      - 8.2.3. SystemImager
        
        8.2.3.1. Image server setup
        
        8.2.3.2. Golden client setup
        
        8.2.3.3. Retrieving the image
        
        8.2.3.4. Cloning the systems
        
        8.2.3.5. Other tasks
    - 8.3. Notes for OSCAR and Rocks Users
  - 9. Programming Software
    - 9.1. Programming Languages
    - 9.2. Selecting a Library
    - 9.3. LAM/MPI
      - 9.3.1. Installing LAM/MPI
      - 9.3.2. User Configuration
      - 9.3.3. Using LAM/MPI
      - 9.3.4. Testing the Installation
    - 9.4. MPICH
      - 9.4.1. Installing
      - 9.4.2. User Configuration
      - 9.4.3. Using MPICH
      - 9.4.4. Testing the Installation
      - 9.4.5. MPE
    - 9.5. Other Programming Software
      - 9.5.1. Debuggers
      - 9.5.2. HDF5
      - 9.5.3. SPRNG
    - 9.6. Notes for OSCAR Users
      - 9.6.1. Adding MPE
    - 9.7. Notes for Rocks Users
  - 10. Management Software
    - 10.1. C3
      - 10.1.1. Installing C3
      - 10.1.2. Using C3 Commands
        
        10.1.2.1. cexec
        
        10.1.2.2. cget
        
        10.1.2.3. ckill
        
        10.1.2.4. cpush
        
        10.1.2.5. crm
        
        10.1.2.6. cshutdown
        
        10.1.2.7. clist, cname, and cnum
        
        10.1.2.8. Further examples and comments
    - 10.2. Ganglia
      - 10.2.1. Installing and Using Ganglia
        
        10.2.1.1. RRDTool
        
        10.2.1.2. Apache and PHP
        
        10.2.1.3. Ganglia monitor core
        
        10.2.1.4. Web frontend
    - 10.3. Notes for OSCAR and Rocks Users
  - 11. Scheduling Software
    - 11.1. OpenPBS
      - 11.1.1. Architecture
      - 11.1.2. Installing OpenPBS
      - 11.1.3. Configuring PBS
      - 11.1.4. Managing PBS
      - 11.1.5. Using PBS
      - 11.1.6. PBS's GUI
      - 11.1.7. Maui Scheduler
    - 11.2. Notes for OSCAR and Rocks Users
  - 12. Parallel Filesystems
    - 12.1. PVFS
      - 12.1.1. Installing PVFS on the Head Node
      - 12.1.2. Configuring the Metadata Server
      - 12.1.3. I/O Server Setup
      - 12.1.4. Client Setup
      - 12.1.5. Running PVFS
        
        12.1.5.1. Troubleshooting
    - 12.2. Using PVFS
    - 12.3. Notes for OSCAR and Rocks Users
- IV. Cluster Programming
  - 13. Getting Started with MPI
    - 13.1. MPI
      - 13.1.1. Core MPI
        
        13.1.1.1. MPI_Init
        
        13.1.1.2. MPI_Finalize
        
        13.1.1.3. MPI_Comm_size
        
        13.1.1.4. MPI_Comm_rank
        
        13.1.1.5. MPI_Get_processor_name
    - 13.2. A Simple Problem
      - 13.2.1. Background
      - 13.2.2. Single-Processor Program
    - 13.3. An MPI Solution
      - 13.3.1. A C Solution
      - 13.3.2. Transferring Data
        
        13.3.2.1. MPI_Send
        
        13.3.2.2. MPI_Recv
      - 13.3.3. MPI Using FORTRAN
      - 13.3.4. MPI Using C++
    - 13.4. I/O with MPI
    - 13.5. Broadcast Communications
      - 13.5.1. Broadcast Functions
        
        13.5.1.1. MPI_Bcast
        
        13.5.1.2. MPI_Reduce
  - 14. Additional MPI Features
    - 14.1. More on Point-to-Point Communication
      - 14.1.1. Non-Blocking Communication
        
        14.1.1.1. MPI_Isend and MPI_Irecv
        
        14.1.1.2. MPI_Wait
        
        14.1.1.3. MPI_Test
        
        14.1.1.4. MPI_Iprobe
        
        14.1.1.5. MPI_Cancel
        
        14.1.1.6. MPI_Sendrecv and MPI_Sendrecv_replace
    - 14.2. More on Collective Communication
      - 14.2.1. Gather and Scatter
        
        14.2.1.1. MPI_Gather
        
        14.2.1.2. MPI_Scatter
    - 14.3. Managing Communicators
      - 14.3.1. Communicator Commands
        
        14.3.1.1. MPI_Comm_group
        
        14.3.1.2. MPI_Group_incl and MPI_Group_excl
        
        14.3.1.3. MPI_Comm_create
        
        14.3.1.4. MPI_Comm_free and MPI_Group_free
        
        14.3.1.5. MPI_Comm_split
    - 14.4. Packaging Data
      - 14.4.1. User-Defined Types
        
        14.4.1.1. MPI_Type_struct
        
        14.4.1.2. MPI_Type_commit
      - 14.4.2. Packing Data
        
        14.4.2.1. MPI_Pack
        
        14.4.2.2. MPI_Unpack
  - 15. Designing Parallel Programs
    - 15.1. Overview
    - 15.2. Problem Decomposition
      - 15.2.1. Decomposition Strategies
        
        15.2.1.1. Data decomposition
        
        15.2.1.2. Control decomposition
    - 15.3. Mapping Tasks to Processors
      - 15.3.1. Communication Overhead
      - 15.3.2. Load Balancing
    - 15.4. Other Considerations
      - 15.4.1. Parallel I/O
      - 15.4.2. MPI-IO Functions
        
        15.4.2.1. MPI_File_open
        
        15.4.2.2. MPI_File_seek
        
        15.4.2.3. MPI_File_read
        
        15.4.2.4. MPI_File_close
      - 15.4.3. Random Numbers
  - 16. Debugging Parallel Programs
    - 16.1. Debugging and Parallel Programs
    - 16.2. Avoiding Problems
    - 16.3. Programming Tools
    - 16.4. Rereading Code
    - 16.5. Tracing with printf
    - 16.6. Symbolic Debuggers
      - 16.6.1. gdb
      - 16.6.2. ddd
    - 16.7. Using gdb and ddd with MPI
    - 16.8. Notes for OSCAR and Rocks Users
  - 17. Profiling Parallel Programs
    - 17.1. Why Profile?
    - 17.2. Writing and Optimizing Code
    - 17.3. Timing Complete Programs
    - 17.4. Timing C Code Segments
      - 17.4.1. Manual Timing with MPI
      - 17.4.2. MPI Functions
        
        17.4.2.1. MPI_Wtime
        
        17.4.2.2. MPI_Wtick
        
        17.4.2.3. MPI_Barrier
      - 17.4.3. PMPI
    - 17.5. Profilers
      - 17.5.1. gprof
      - 17.5.2. gcov
      - 17.5.3. Profiling Parallel Programs with gprof and gcov
    - 17.6. MPE
      - 17.6.1. Using MPE
    - 17.7. Customized MPE Logging
    - 17.8. Notes for OSCAR and Rocks Users
- V. Appendix
  - A. References
    - A.1. Books
    - A.2. URLs
      - A.2.1. General Cluster Information
      - A.2.2. Linux
      - A.2.3. Cluster Software
      - A.2.4. Grid Computing and Tools
      - A.2.5. Cloning and Management Software
      - A.2.6. Filesystems
      - A.2.7. Parallel Benchmarks
      - A.2.8. Programming Software
      - A.2.9. Scheduling Software
      - A.2.10. System Software and Utilities
- About the Author
- Colophon
- SPECIAL OFFER: Upgrade this ebook with OReilly