IBM TotalStorage DS6000 series : performance monitoring and tuning / / Cathy Warrick ... [et al.]

• Pubblicazione: San Jose, CA, : IBM, International Technical Support Organization, c2005
• Edizione: 1st ed.
• Descrizione fisica: xxii, 507 p. : ill
• Disciplina: 004.6
• Soggetto topico: Storage area networks (Computer networks); Computer storage devices; System design; IBM computers
• Persona (resp. second.): WarrickCathy
• Note generali: "December 2005."
• Nota di bibliografia: Includes bibliographical references and index.
• Nota di contenuto: Front cover -- Contents -- Figures -- Notices -- Trademarks -- Preface -- The team that wrote this redbook -- Become a published author -- Comments welcome -- Chapter 1. Model characteristics -- 1.1 Benefits of the DS6000 series -- 1.1.1 Infrastructure Simplification -- 1.1.2 Business Continuity -- 1.1.3 Information Life Cycle Management -- 1.2 Hardware overview -- 1.2.1 DS6800 controller enclosure (Model 1750-511) -- 1.2.2 DS6000 expansion enclosure (Model 1750-EX1) -- 1.3 Functional overview -- 1.3.1 Storage capacity: RAID 5 and RAID 10 -- 1.3.2 Resiliency -- 1.3.3 Copy Services -- 1.4 DS management tools -- 1.4.1 IBM TotalStorage DS Storage Manager -- 1.4.2 IBM TotalStorage DS Open API -- 1.4.3 IBM TotalStorage DS Command Line Interface (CLI) -- 1.5 Supported environments -- 1.6 Performance overview -- 1.6.1 Sequential Prefetching in Adaptive Replacement Cache (SARC) -- 1.6.2 Performance for zSeries -- 1.6.3 IBM TotalStorage Multi-path Subsystem Device Driver (SDD) -- 1.6.4 Command Tag Queuing -- 1.7 IBM TotalStorage DS family comparisons -- 1.7.1 DS6000 series compared to ESS -- 1.7.2 DS6000 series compared to DS8000 -- 1.7.3 DS6000 series compared to DS4000 series -- 1.8 IBM DS6000 combined with virtualization products -- 1.8.1 IBM SAN Volume Controller -- 1.8.2 IBM SAN File System -- Chapter 2. Hardware configuration planning -- 2.1 Rules of thumb and benchmarks -- 2.2 Understanding your workload characteristics -- 2.3 DS6000 major hardware components -- 2.4 DS6000 server processor -- 2.5 Cache and persistent memory (formerly NVS) -- 2.5.1 Cache -- 2.5.2 Persistent memory -- 2.5.3 Cache algorithms -- 2.5.4 Cache size consideration -- 2.6 DS6000 disks -- 2.6.1 DS6000 disk capacity -- 2.6.2 Disk four-packs -- 2.6.3 Disk four-pack capacity -- 2.6.4 Disk four-pack intermixing -- 2.6.5 Disk conversions -- 2.7 Choosing the DS6000 disks. ; 2.7.1 Disk capacity -- 2.7.2 Disk Magic examples using 146 GB and 300 GB disk drives -- 2.7.3 Disk speed (RPM) -- 2.7.4 Disk Magic examples using 15K rpm and 10K rpm disk drives -- 2.8 RAID implementation -- 2.8.1 RAID Arrays -- 2.8.2 RAID 5 array -- 2.8.3 RAID 10 array -- 2.8.4 Combination of RAID 5 and RAID 10 arrays -- 2.8.5 RAID 5 versus RAID 10 performance -- 2.9 SBOD (Switched Bunch of Disks) connection -- 2.9.1 Standard storage subsystem FC-AL problem -- 2.9.2 Switched FC-AL advantages -- 2.9.3 DS6000 switched FC-AL implementation -- 2.10 Host adapter -- 2.10.1 Host adapter configuration -- 2.10.2 FCP attachment -- 2.10.3 FICON attachment -- 2.10.4 Preferred Path -- 2.11 Tools to aid in hardware planning -- 2.11.1 Whitepapers -- 2.11.2 Disk Magic -- 2.11.3 Capacity Magic -- Chapter 3. Logical configuration planning -- 3.1 Principles for performance optimization: balance, isolation and spread -- 3.1.1 Isolation -- 3.1.2 Resource sharing -- 3.1.3 Spreading -- 3.1.4 Using isolation, resource-sharing and spreading to optimize the DS6000 performance -- 3.2 Isolation requirements -- 3.2.1 Review the application workload characteristics to determine the isolation requirements -- 3.3 Plan assignment of DS6000 hardware resources to workloads -- 3.3.1 Plan DS6000 hardware resources for isolated workloads -- 3.3.2 Plan DS6000 hardware resources for resource-sharing workloads -- 3.3.3 Spread volumes and host connections across available hardware -- 3.4 Logical configuration - components and terminology -- 3.5 Configuring for performance -- 3.5.1 Mixing drive geometries -- 3.5.2 Mixing open and zSeries logical disks -- 3.5.3 Arrays and Array Sites -- 3.5.4 Select a Rank format -- 3.5.5 Extent Pool implications -- 3.5.6 Number of Ranks in an Extent Pool -- 3.5.7 LSS design -- 3.5.8 Preferred paths -- 3.6 Performance and sizing considerations. ; 3.6.1 Workload characteristics -- 3.6.2 Data placement in the DS6000 -- 3.6.3 Open systems LVM striping -- 3.7 .Performance and sizing considerations for z/OS -- 3.7.1 Performance potential in z/OS environments -- 3.8 Logical disks - number and size -- 3.9 Logical disk sizes - general considerations -- 3.9.1 Future requirements -- 3.9.2 Maximum number of devices -- 3.10 Configuring I/O ports -- 3.10.1 Multiple host attachment -- Chapter 4. Planning and monitoring tools -- 4.1 Disk Magic -- 4.1.1 Overview and characteristics -- 4.1.2 Output information -- 4.1.3 Disk Magic modeling -- 4.1.4 Disk Magic for zSeries -- 4.1.5 Disk Magic for open systems -- 4.1.6 Workload growth projection -- 4.1.7 Input data needed for DIsk Magic study -- 4.2 Capacity Magic -- 4.2.1 Overview and features -- 4.2.2 Wizard -- 4.2.3 Graphical interface -- 4.2.4 Reports -- 4.2.5 Examples -- 4.3 IBM TotalStorage Productivity Center for Disk -- 4.3.1 IBM TotalStorage Productivity Center -- 4.3.2 IBM TotalStorage Productivity Center for Disk -- 4.3.3 Operation characteristics -- 4.3.4 Using IBM TotalStorage Productivity Center for Disk -- 4.3.5 Exploiting gauges -- 4.3.6 Interpreting the DS6000 performance -- 4.3.7 Performance gauge - considerations -- 4.3.8 IBM TotalStorage Productivity Center for Disk and other tools -- 4.3.9 IBM TotalStorage Productivity Center for Disk in mixed environment -- 4.4 SAN statistics -- 4.5 Monitoring performance through a SAN switch or director -- Chapter 5. Host attachment -- 5.1 DS6000 host attachment -- 5.1.1 Attaching to open systems hosts -- 5.1.2 FICON-attached S/390 and zSeries hosts -- 5.1.3 Example of host attachments -- 5.2 Multipathing -- 5.3 FICON -- 5.4 Fibre Channel -- 5.4.1 Supported Fibre Channel attached hosts -- 5.4.2 Fibre Channel topologies -- 5.5 SAN implementations -- 5.5.1 Description and characteristics of a SAN. ; 5.5.2 Benefits of a SAN -- 5.5.3 SAN cabling for availability and performance -- 5.5.4 Importance of establishing zones -- 5.5.5 LUN masking -- 5.5.6 Configuring logical disks in a SAN -- 5.6 Subsystem Device Driver (SDD) - multipathing -- 5.6.1 SDD load balancing -- 5.6.2 Concurrent LMC load -- 5.6.3 Single path mode -- 5.6.4 Single FC adapter with multiple paths -- 5.6.5 Path failover and online recovery -- 5.6.6 Using SDDPCM on an AIX host system -- 5.6.7 SDD datapath command -- Chapter 6. IBM TotalStorage SAN Volume Controller attachment -- 6.1 IBM TotalStorage SAN Volume Controller -- 6.1.1 IBM TotalStorage SAN Volume Controller concepts -- 6.1.2 SAN Volume Controller multipathing -- 6.1.3 Copy Services -- 6.1.4 SAN Volume Controller performance considerations -- 6.2 DS6000 performance considerations -- 6.2.1 DS6000 Array -- 6.2.2 DS6000 Rank format -- 6.2.3 DS6000 Extent Pool implications -- 6.2.4 DS6000 volumes consideration -- 6.2.5 Volume assignment to SAN Volume Controller cluster -- 6.2.6 Number of paths to attach the DS6000 to SAN Volume Controller -- 6.3 Performance monitoring -- 6.3.1 IBM TotalStorage Productivity Center for Disk -- 6.3.2 Using IBM TotalStorage Productivity Center for Disk to monitor the SAN Volume Controller -- 6.4 Sharing the DS6000 between a host and the IBM SAN Volume Controller -- 6.4.1 Sharing the DS6000 between open systems server hosts and the IBM SAN Volume Controller -- 6.4.2 Sharing the DS6000 between iSeries host and the IBM SAN Volume Controller -- 6.4.3 Sharing the DS6000 between zSeries server host and the IBM SAN Volume Controller -- 6.5 Advanced functions for the DS6000 -- 6.6 Volume creation and deletion on the DS6000 -- 6.7 Configuration guidelines -- Chapter 7. Open systems servers - UNIX -- 7.1 UNIX performance monitoring and tuning -- 7.2 Planning and preparing UNIX servers for performance. ; 7.2.1 I/O balanced across Extent Pools -- 7.2.2 DS6000 LUN size -- 7.2.3 Document the LUN assignments -- 7.2.4 Multipathing considerations -- 7.2.5 System and adapter code level -- 7.3 Common UNIX performance monitoring tools -- 7.3.1 iostat -- 7.3.2 SAR -- 7.3.3 vmstat -- 7.4 AIX-specific I/O monitoring commands and tools -- 7.4.1 topas -- 7.4.2 nmon -- 7.4.3 filemon -- 7.4.4 lvmstat -- 7.5 HP-UX specific I/O monitoring commands -- 7.6 SDD commands for AIX, HP-UX, and Solaris -- 7.6.1 HP-UX SDD commands -- 7.6.2 Sun Solaris SDD commands -- 7.7 Testing and verifying DS6000 storage -- 7.7.1 Using the dd command to test sequential Rank reads and writes -- 7.7.2 Verifying your system -- 7.8 Volume groups, logical volumes and file systems -- 7.8.1 Creating the volume group -- 7.8.2 Creating a logical volume -- 7.8.3 Creating the file system -- 7.9 Operating system tuning -- 7.9.1 AIX operating system tuning (JFS and JFS2) -- 7.9.2 HP-UX OS tuning for sequential I/O -- 7.9.3 Sun Solaris OS tuning for sequential I/O -- Chapter 8. Open system servers - Linux for xSeries -- 8.1 Supported Linux distributions -- 8.2 Introduction to Linux OS components -- 8.2.1 Understanding and tuning virtual memory -- 8.2.2 Understanding and tuning the swap partition -- 8.2.3 Understanding and tuning the daemons -- 8.2.4 Compiling the kernel -- 8.2.5 Changing kernel parameters -- 8.2.6 Kernel parameters -- 8.2.7 Understanding and tuning the file systems -- 8.2.8 Tuning TCP window size -- 8.3 Linux monitoring tools -- 8.3.1 uptime -- 8.3.2 dmesg -- 8.3.3 top -- 8.3.4 iostat -- 8.3.5 vmstat -- 8.3.6 sar -- 8.3.7 isag -- 8.3.8 GKrellM -- 8.3.9 KDE System Guard -- 8.4 Host bus adapter (HBA) settings -- 8.5 Logical Volume Manager for Linux (LVM) -- 8.5.1 Implementation -- 8.5.2 Performance management -- 8.6 Bonnie -- 8.6.1 Benchmarks -- 8.6.2 Downloading -- 8.7 Bonnie++. ; 8.8 Disk bottlenecks.
• Sommario/riassunto: This IBM Redbooks publication provides guidance about how to configure, monitor, and manage your IBM TotalStorage DS6000 to achieve optimum performance. We describe the DS6000 performance features and characteristics and how they can be exploited with the different server platforms that can attach to it. Then in consecutive chapters we detail the specific performance recommendations and discussions that apply for each server environment, as well as for database and Copy Services environments. We also outline the various tools available for monitoring and measuring I/O performance for the different server environments, as well as how to monitor performance of the entire DS6000 subsystem.
• Titolo autorizzato: IBM TotalStorage DS6000 series
• Formato: Materiale a stampa
• Livello bibliografico: Monografia
• Lingua di pubblicazione: Inglese
• Record Nr.: 9910810921103321
• Lo trovi qui: Univ. Federico II
• Serie: IBM redbooks.