IBM e server pSeries Cluster systems handbook

The IBM eServer Cluster 1600 server, which was introduced to meet the rigorous demands of mission-critical enterprise applications, continues to offer outstanding performance, scalability, reliability, availability, serviceability, and management capabilities. In this IBM Redbooks publication, we hi...

Descripción completa

Detalles Bibliográficos
Autor Corporativo: International Business Machines Corporation. International Technical Support Organization (-)
Otros Autores: Quintero, Dino (-)
Formato: Libro electrónico
Idioma:Inglés
Publicado: [S.l.] : IBM, International Technical Support Organization c2003.
Edición:1st ed
Colección:IBM redbooks.
Materias:
Ver en Biblioteca Universitat Ramon Llull:https://discovery.url.edu/permalink/34CSUC_URL/1im36ta/alma991009627161806719
Tabla de Contenidos:
  • Front cover
  • Contents
  • Notices
  • Trademarks
  • Preface
  • Become a published author
  • Comments welcome
  • Chapter 1. Introduction
  • 1.1 Overview of Cluster 1600
  • 1.2 Choosing PSSP or CSM
  • 1.2.1 Cluster management with PSSP
  • 1.2.2 Cluster management with CSM
  • Chapter 2. Cluster 1600 hardware
  • 2.1 Overview
  • 2.2 Cluster 1600 hardware components
  • 2.2.1 Nodes
  • 2.2.2 Frames
  • 2.2.3 Switches
  • 2.2.4 PSSP control workstations
  • 2.2.5 CSM for AIX management server
  • 2.2.6 Hardware Management Console (HMC)
  • 2.3 CSM and PSSP hardware support
  • 2.3.1 CSM-managed node requirements
  • 2.4 PSSP control workstation
  • 2.4.1 Control workstation requirements
  • 2.4.2 Supported control workstations
  • 2.4.3 High Availability Control Workstation
  • 2.4.4 HACWS limitations
  • 2.5 CSM management server
  • 2.5.1 Memory and disk space
  • 2.5.2 Network requirements
  • 2.5.3 Asynchronous card requirements
  • 2.5.4 Using a Logical Partition (LPAR) as a CSM management server
  • 2.6 Cluster 1600 server concepts
  • 2.6.1 pSeries architecture
  • 2.6.2 Cluster 1600 and the HMC
  • 2.6.3 Firmware
  • 2.6.4 Electronic Service Agent
  • 2.6.5 Planning for Cluster 1600 servers
  • 2.7 Hardware supported and currently marketed
  • 2.8 pSeries servers
  • 2.8.1 pSeries 615 server (7029-6C3 and 6E3 deskside)
  • 2.8.2 pSeries 630 server (7028-6C4 and 6E4 deskside)
  • 2.8.3 pSeries 650 server (7038-6M2)
  • 2.8.4 pSeries 655 server (7039-651)
  • 2.8.5 670 Server (7040-671)
  • 2.8.6 690 server (7040-681)
  • 2.8.7 xSeries servers
  • 2.9 Switches
  • 2.9.1 9076 model 555
  • 2.9.2 9076 model 556
  • 2.9.3 9076 model 557
  • 2.9.4 9076 model 558
  • 2.9.5 7045-SW4 pSeries HPS (High Performance Switch)
  • 2.10 Switch adapters
  • 2.10.1 Switch adapter placement restrictions
  • 2.10.2 pSeries HPS switch network interface cards (SNI).
  • 2.11 Legacy hardware supported but no longer marketed
  • 2.11.1 pSeries 660 Model 6M1 (7026-6M1)
  • Chapter 3. Network configuration
  • 3.1 SP LAN Ethernet
  • 3.1.1 Supported Ethernet adapters and their placement
  • 3.1.2 Ethernet network topology
  • 3.1.3 IP label convention
  • 3.2 Switch network
  • 3.2.1 Benefits of a Switch network
  • 3.2.2 SP Switch2
  • 3.2.3 Switch IP network and addressing
  • 3.3 Other networks
  • 3.4 Network considerations
  • 3.4.1 The RS-232 connection
  • 3.4.2 System topology considerations
  • 3.4.3 Boot/install server requirements
  • 3.4.4 The SP Ethernet administrative LAN
  • 3.4.5 Additional LANs - considerations
  • 3.4.6 IP over the switch - considerations
  • 3.4.7 Subnetting - considerations
  • 3.4.8 HMC trusted network - considerations
  • 3.4.9 Network router node considerations
  • 3.4.10 Clustered server configuration considerations
  • 3.4.11 SP-attached server considerations
  • 3.5 Sample scenarios of Cluster 1600 managed by PSSP
  • 3.5.1 CWS with two HMCs and four pSeries
  • 3.5.2 One CWS, one HMC and one pSeries
  • 3.5.3 CWS, two HMCs, two 9076s, with one pSeries and SP Switch2
  • 3.5.4 CWS, HMC, 9076 frame and pSeries with SP Switch
  • 3.6 Networking for Cluster Systems Management (CSM)
  • 3.6.1 CSM hardware control
  • 3.6.2 Hardware and network requirements
  • 3.6.3 Virtual LANs (VLANs)
  • 3.6.4 Conceptual diagram for pSeries cluster
  • 3.6.5 pSeries HPS switch network overview
  • 3.6.6 Switch Network Manager (SNM)
  • 3.6.7 Considerations for Cluster 1600 managed by CSM network
  • 3.6.8 Examples
  • 3.6.9 Redundant HMC Layout for pseries HPS in Cluster 1600
  • 3.6.10 Redundant layout for pSeries in the Cluster 1600
  • 3.6.11 Conceptual Cluster 1600 without a pSeries HPS
  • 3.6.12 Management Server with two HMCs and four pSeries
  • Chapter 4. Software support
  • 4.1 Software components of the Cluster 1600.
  • 4.2 Parallel System Support Programs (PSSP)
  • 4.2.1 Administration and operation
  • 4.2.2 Reliable Scalable Cluster Technology (RSCT)
  • 4.2.3 IBM Virtual Shared Disk (VSD)
  • 4.2.4 Security
  • 4.2.5 Communication subsystem
  • 4.2.6 Network Time Protocol (NTP)
  • 4.2.7 System availability
  • 4.2.8 Other PSSP services
  • 4.2.9 New in PSSP 3.5
  • 4.2.10 Software requirements
  • 4.2.11 Software compatibility matrix
  • 4.2.12 Documentation references - PSSP
  • 4.3 Cluster Systems Management (CSM)
  • 4.3.1 Administration and operation
  • 4.3.2 Reliable Scalable Cluster Technology (RSCT)
  • 4.3.3 New in CSM 1.3.2 for AIX
  • 4.3.4 Supported platform
  • 4.3.5 PSSP-to-CSM transition
  • 4.3.6 Documentation references - CSM
  • 4.4 General Parallel File System (GPFS)
  • 4.4.1 Architecture
  • 4.4.2 Administration and operation
  • 4.4.3 Higher performance/scalability
  • 4.4.4 Recoverability
  • 4.4.5 Migration
  • 4.4.6 New in GPFS 2.1 for AIX 5L
  • 4.4.7 Software requirements
  • 4.4.8 Documentation references
  • 4.5 LoadLeveler
  • 4.5.1 Administration and operations
  • 4.5.2 Capabilities
  • 4.5.3 New in LoadLeveler 3.2
  • 4.5.4 New in LoadLeveler 3.1
  • 4.5.5 Software requirement
  • 4.5.6 LoadLeveler configuration suggestions
  • 4.5.7 Documentation references - LoadLeveler
  • 4.6 Scientific subroutine libraries
  • 4.6.1 Engineering and Scientific Subroutines Library (ESSL) family of products
  • 4.6.2 Operations
  • 4.6.3 New in ESSL 4.1
  • 4.6.4 New in Parallel ESSL 3.1
  • 4.6.5 Software requirements
  • 4.6.6 Documentation references - ESSL and PESSL
  • 4.6.7 Mathematical Acceleration Subsystem (MASS)
  • 4.7 Parallel Environment (PE)
  • 4.7.1 Parallel Programming support
  • 4.7.2 Operation
  • 4.7.3 New in PE 4.1
  • 4.7.4 Software requirements
  • 4.7.5 Documentation references - Parallel Environment (PE).
  • 4.8 IBM High Availability Cluster Multi-Processing for AIX (HACMP)
  • 4.8.1 HACMP operations
  • 4.8.2 Administration and operation
  • 4.8.3 New in HACMP 5.1
  • 4.8.4 Software requirements
  • 4.8.5 Documentation references - HACMP
  • 4.9 Performance Toolbox (PTX) and Performance AIDE (PAIDE)
  • 4.9.1 Administration and operation
  • 4.9.2 Platform requirements
  • 4.10 Software ordering and configuration
  • Chapter 5. Solutions and offerings "best practices"
  • 5.1 High Performance Computing (HPC) environment
  • 5.1.1 A hypothetical solution
  • 5.1.2 Solution architecture
  • 5.1.3 Solution discussion
  • 5.1.4 Cluster management considerations
  • 5.2 Transition from SP nodes to LPARs
  • 5.2.1 System migration by utilizing alternate disk migration
  • 5.3 Virtual Serial port implications with LPARS
  • 5.3.1 Console device
  • 5.3.2 Serial port implication in an LPAR/SP environment
  • 5.3.3 Virtual terminal window
  • 5.4 HMC considerations
  • 5.4.1 Redundant HMC
  • 5.5 Web-based System Manager client solutions
  • 5.5.1 Web-based System Manager functionality through the firewall
  • Appendix A. Performance
  • A.1 Switch performance
  • A.2 Adapter performance
  • A.3 Node I/O slot performance
  • A.4 Application performance
  • A.5 MPI/user space
  • A.6 TCP/IP
  • Related publications
  • IBM Redbooks
  • Other publications
  • Online resources
  • How to get IBM Redbooks
  • Help from IBM
  • Abbreviations and acronyms
  • Index
  • Back cover.