book

IBM System Storage SAN Volume Controller Best Practices and Performance Guidelines

by Mary Lovelace, Katja Gebuhr, Ivo Gomilsek, Ronda Hruby, Paulo Neto, Jon Parkes, Otavio Rocha Filho, Leandro Torolho

December 2012

Intermediate to advanced

516 pages

15h 6m

English

IBM Redbooks

Read now

Unlock full access

Trademarks
The team who wrote this bookNow you can become a published author, too!Comments welcomeStay connected to IBM Redbooks
December 2012, Third EditionDecember 2008, Second Edition
1.1 Enhancements and changes in SAN Volume Controller V5.11.2 Enhancements and changes in SAN Volume Controller V6.11.3 Enhancements and changes in SAN Volume Controller V6.2
2.1 SAN topology of the SAN Volume Controller2.1.1 Redundancy2.1.2 Topology basics2.1.3 ISL oversubscription2.1.4 Single switch SAN Volume Controller SANs2.1.5 Basic core-edge topology2.1.6 Four-SAN, core-edge topology2.1.7 Common topology issues2.1.8 Split clustered system or stretch clustered system2.2 SAN switches2.2.1 Selecting SAN switch models2.2.2 Switch port layout for large SAN edge switches2.2.3 Switch port layout for director-class SAN switches2.2.4 IBM System Storage and Brocade b-type SANs2.2.5 IBM System Storage and Cisco SANs2.2.6 SAN routing and duplicate worldwide node names2.3 Zoning2.3.1 Types of zoning2.3.2 Prezoning tips and shortcuts2.3.3 SAN Volume Controller internode communications zone2.3.4 SAN Volume Controller storage zones2.3.5 SAN Volume Controller host zones2.3.6 Standard SAN Volume Controller zoning configuration2.3.7 Zoning with multiple SAN Volume Controller clustered systems2.3.8 Split storage subsystem configurations2.4 Switch domain IDs2.5 Distance extension for remote copy services2.5.1 Optical multiplexors2.5.2 Long-distance SFPs or XFPs2.5.3 Fibre Channel IP conversion2.6 Tape and disk traffic that share the SAN2.7 Switch interoperability2.8 IBM Tivoli Storage Productivity Center2.9 iSCSI support2.9.1 iSCSI initiators and targets2.9.2 iSCSI Ethernet configuration2.9.3 Security and performance2.9.4 Failover of port IP addresses and iSCSI names2.9.5 iSCSI protocol limitations
3.1 Advantages of virtualization3.1.1 Features of the SAN Volume Controller3.2 Scalability of SAN Volume Controller clustered systems3.2.1 Advantage of multiclustered systems versus single-clustered systems3.2.2 Growing or splitting SAN Volume Controller clustered systems3.2.3 Adding or upgrading SVC node hardware3.3 Clustered system upgrade
4.1 Controller affinity and preferred path4.2 Considerations for DS4000 and DS50004.2.1 Setting the DS4000 and DS5000 so that both controllers have the same worldwide node name4.2.2 Balancing workload across DS4000 and DS5000 controllers4.2.3 Ensuring path balance before MDisk discovery4.2.4 Auto-Logical Drive Transfer for the DS4000 and DS50004.2.5 Selecting array and cache parameters4.2.6 Logical drive mapping4.3 Considerations for DS80004.3.1 Balancing workload across DS8000 controllers4.3.2 DS8000 ranks to extent pools mapping4.3.3 Mixing array sizes within a storage pool4.3.4 Determining the number of controller ports for the DS80004.3.5 LUN masking4.3.6 WWPN to physical port translation4.4 Considerations for IBM XIV Storage System4.4.1 Cabling considerations4.4.2 Host options and settings for XIV systems4.4.3 Restrictions4.5 Considerations for IBM Storwize V70004.5.1 Defining internal storage4.5.2 Configuring Storwize V7000 storage systems4.6 Considerations for third-party storage: EMC SymmetrixDMX and Hitachi Data Systems4.7 Medium error logging4.8 Mapping physical LBAs to volume extents4.9 Identifying storage controller boundaries with IBM Tivoli Storage Productivity Center
5.1 Availability considerations for storage pools5.2 Selecting storage subsystems5.3 Selecting the storage pool5.3.1 Selecting the number of arrays per storage pool5.3.2 Selecting LUN attributes5.3.3 Considerations for the IBM XIV Storage System5.4 Quorum disk considerations for SAN Volume Controller5.5 Tiered storage5.6 Adding MDisks to existing storage pools5.6.1 Checking access to new MDisks5.6.2 Persistent reserve5.6.3 Renaming MDisks5.7 Restriping (balancing) extents across a storage pool5.7.1 Installing prerequisites and the SVCTools package5.7.2 Running the extent balancing script5.8 Removing MDisks from existing storage pools5.8.1 Migrating extents from the MDisk to be deleted5.8.2 Verifying the identity of an MDisk before removal5.8.3 Correlating the back-end volume (LUN) with the MDisk5.9 Remapping managed MDisks5.10 Controlling extent allocation order for volume creation5.11 Moving an MDisk between SVC clusters

6.1 Overview of volumes6.1.1 Striping compared to sequential type6.1.2 Thin-provisioned volumes6.1.3 Space allocation6.1.4 Thin-provisioned volume performance6.1.5 Limits on virtual capacity of thin-provisioned volumes6.1.6 Testing an application with a thin-provisioned volume6.2 Volume mirroring6.2.1 Creating or adding a mirrored volume6.2.2 Availability of mirrored volumes6.2.3 Mirroring between controllers6.3 Creating volumes6.3.1 Selecting the storage pool6.3.2 Changing the preferred node within an I/O group6.3.3 Moving a volume to another I/O group6.4 Volume migration6.4.1 Image-type to striped-type migration6.4.2 Migrating to image-type volume6.4.3 Migrating with volume mirroring6.5 Preferred paths to a volume6.5.1 Governing of volumes6.6 Cache mode and cache-disabled volumes6.6.1 Underlying controller remote copy with SAN Volume Controller cache-disabled volumes6.6.2 Using underlying controller FlashCopy with SAN Volume Controller cache disabled volumes6.6.3 Changing the cache mode of a volume6.7 Effect of a load on storage controllers6.8 Setting up FlashCopy services6.8.1 Making a FlashCopy volume with application data integrity6.8.2 Making multiple related FlashCopy volumes with data integrity6.8.3 Creating multiple identical copies of a volume6.8.4 Creating a FlashCopy mapping with the incremental flag6.8.5 Using thin-provisioned FlashCopy6.8.6 Using FlashCopy with your backup application6.8.7 Migrating data by using FlashCopy6.8.8 Summary of FlashCopy rules6.8.9 IBM Tivoli Storage FlashCopy Manager6.8.10 IBM System Storage Support for Microsoft Volume Shadow Copy Service
7.1 Introduction to remote copy services7.1.1 Common terminology and definitions7.1.2 Intercluster link7.2 SAN Volume Controller remote copy functions by release7.2.1 Remote copy in SAN Volume Controller V6.27.2.2 Remote copy features by release7.3 Terminology and functional concepts7.3.1 Remote copy partnerships and relationships7.3.2 Global Mirror control parameters7.3.3 Global Mirror partnerships and relationships7.3.4 Asynchronous remote copy7.3.5 Understanding remote copy write operations7.3.6 Asynchronous remote copy7.3.7 Global Mirror write sequence7.3.8 Write ordering7.3.9 Colliding writes7.3.10 Link speed, latency, and bandwidth7.3.11 Choosing a link cable of supporting Global Mirror applications7.3.12 Remote copy volumes: Copy directions and default roles7.4 Intercluster link7.4.1 SAN configuration overview7.4.2 Switches and ISL oversubscription7.4.3 Zoning7.4.4 Distance extensions for the intercluster link7.4.5 Optical multiplexors7.4.6 Long-distance SFPs and XFPs7.4.7 Fibre Channel IP conversion7.4.8 Configuration of intercluster links7.4.9 Link quality7.4.10 Hops7.4.11 Buffer credits7.5 Global Mirror design points7.5.1 Global Mirror parameters7.5.2 The chcluster and chpartnership commands7.5.3 Distribution of Global Mirror bandwidth7.5.4 1920 errors7.6 Global Mirror planning7.6.1 Rules for using Metro Mirror and Global Mirror7.6.2 Planning overview7.6.3 Planning specifics7.7 Global Mirror use cases7.7.1 Synchronizing a remote copy relationship7.7.2 Setting up Global Mirror relationships, saving bandwidth, and resizing volumes7.7.3 Master and auxiliary volumes and switching their roles7.7.4 Migrating a Metro Mirror relationship to Global Mirror7.7.5 Multiple cluster mirroring7.7.6 Performing three-way copy service functions7.7.7 When to use storage controller Advanced Copy Services functions7.7.8 Using Metro Mirror or Global Mirror with FlashCopy7.7.9 Global Mirror upgrade scenarios7.8 Intercluster Metro Mirror and Global Mirror source as an FC target7.9 States and steps in the Global Mirror relationship7.9.1 Global Mirror states7.9.2 Disaster recovery and Metro Mirror and Global Mirror states7.9.3 State definitions7.10 1920 errors7.10.1 Diagnosing and fixing 1920 errors7.10.2 Focus areas for 1920 errors7.10.3 Recovery7.10.4 Disabling the glinktolerance feature7.10.5 Cluster error code 1920 checklist for diagnosis7.11 Monitoring remote copy relationships
8.1 Configuration guidelines8.1.1 Host levels and host object name8.1.2 The number of paths8.1.3 Host ports8.1.4 Port masking8.1.5 Host to I/O group mapping8.1.6 Volume size as opposed to quantity8.1.7 Host volume mapping8.1.8 Server adapter layout8.1.9 Availability versus error isolation8.2 Host pathing8.2.1 Preferred path algorithm8.2.2 Path selection8.2.3 Path management8.2.4 Dynamic reconfiguration8.2.5 Volume migration between I/O groups8.3 I/O queues8.3.1 Queue depths8.4 Multipathing software8.5 Host clustering and reserves8.5.1 Clearing reserves8.5.2 SAN Volume Controller MDisk reserves8.6 AIX hosts8.6.1 HBA parameters for performance tuning8.6.2 Configuring for fast fail and dynamic tracking8.6.3 Multipathing8.6.4 SDD8.6.5 SDDPCM8.6.6 SDD compared to SDDPCM8.7 Virtual I/O Server8.7.1 Methods to identify a disk for use as a virtual SCSI disk8.7.2 UDID method for MPIO8.7.3 Backing up the virtual I/O configuration8.8 Windows hosts8.8.1 Clustering and reserves8.8.2 SDD versus SDDDSM8.8.3 Tunable parameters8.8.4 Changing back-end storage LUN mappings dynamically8.8.5 Guidelines for disk alignment by using Windows with SAN Volume Controller volumes8.9 Linux hosts8.9.1 SDD compared to DM-MPIO8.9.2 Tunable parameters8.10 Solaris hosts8.10.1 Solaris MPxIO8.10.2 Symantec Veritas Volume Manager8.10.3 ASL specifics for SAN Volume Controller8.10.4 SDD pass-through multipathing8.10.5 DMP multipathing8.10.6 Troubleshooting configuration issues8.11 VMware server8.11.1 Multipathing solutions supported8.11.2 Multipathing configuration maximums8.12 Mirroring considerations8.12.1 Host-based mirroring8.13 Monitoring8.13.1 Automated path monitoring8.13.2 Load measurement and stress tools
9.1 SAN Volume Controller continuing performance enhancements9.2 Solid State Drives and Easy Tier9.2.1 Internal SSD redundancy9.2.2 Performance scalability and I/O groups9.3 Real Time Performance Monitor
10.1 Workload considerations10.2 Tiering10.3 Storage controller considerations10.3.1 Back-end I/O capacity10.4 Array considerations10.4.1 Selecting the number of LUNs per array10.4.2 Selecting the number of arrays per storage pool10.5 I/O ports, cache, and throughput considerations10.5.1 Back-end queue depth10.5.2 MDisk transfer size10.6 SAN Volume Controller extent size10.7 SAN Volume Controller cache partitioning10.8 IBM DS8000 considerations10.8.1 Volume layout10.8.2 Cache10.8.3 Determining the number of controller ports for DS800010.8.4 Storage pool layout10.8.5 Extent size10.9 IBM XIV considerations10.9.1 LUN size10.9.2 I/O ports10.9.3 Storage pool layout10.9.4 Extent size10.9.5 Additional information10.10 Storwize V7000 considerations10.10.1 Volume setup10.10.2 I/O ports10.10.3 Storage pool layout10.10.4 Extent size10.10.5 Additional information10.11 DS5000 considerations10.11.1 Selecting array and cache parameters10.11.2 Considerations for controller configuration10.11.3 Mixing array sizes within the storage pool10.11.4 Determining the number of controller ports for DS4000
11.1 Overview of Easy Tier11.2 Easy Tier concepts11.2.1 SSD arrays and MDisks11.2.2 Disk tiers11.2.3 Single tier storage pools11.2.4 Multitier storage pools11.2.5 Easy Tier process11.2.6 Easy Tier operating modes11.2.7 Easy Tier activation11.3 Easy Tier implementation considerations11.3.1 Prerequisites11.3.2 Implementation rules11.3.3 Easy Tier limitations11.4 Measuring and activating Easy Tier11.4.1 Measuring by using the Storage Advisor Tool11.5 Activating Easy Tier with the SAN Volume Controller CLI11.5.1 Initial cluster status11.5.2 Turning on Easy Tier evaluation mode11.5.3 Creating a multitier storage pool11.5.4 Setting the disk tier11.5.5 Checking the Easy Tier mode of a volume11.5.6 Final cluster status11.6 Activating Easy Tier with the SAN Volume Controller GUI11.6.1 Setting the disk tier on MDisks11.6.2 Checking Easy Tier status
12.1 Application workloads12.1.1 Transaction-based workloads12.1.2 Throughput-based workloads12.1.3 Storage subsystem considerations12.1.4 Host considerations12.2 Application considerations12.2.1 Transaction environments12.2.2 Throughput environments12.3 Data layout overview12.3.1 Layers of volume abstraction12.3.2 Storage administrator and AIX LVM administrator roles12.3.3 General data layout guidelines12.3.4 Database strip size considerations (throughput workload)12.3.5 LVM volume groups and logical volumes12.4 Database storage12.5 Data layout with the AIX Virtual I/O Server12.5.1 Overview12.5.2 Data layout strategies12.6 Volume size12.7 Failure boundaries
13.1 Analyzing the SAN Volume Controller by using Tivoli Storage Productivity Center13.2 Considerations for performance analysis13.2.1 SAN Volume Controller considerations13.2.2 Storwize V7000 considerations13.3 Top 10 reports for SAN Volume Controller and Storwize V700013.3.1 I/O Group Performance reports (report 1) for SAN Volume Controller and Storwize V700013.3.2 Node Cache Performance reports (report 2) for SAN Volume Controller and Storwize V700013.3.3 Managed Disk Group Performance report (reports 3 and 4) for SAN Volume Controller13.3.4 Top Volume Performance reports (reports 5 - 9) for SAN Volume Controller and Storwize V700013.3.5 Port Performance reports (report 10) for SAN Volume Controller and Storwize V700013.4 Reports for fabric and switches13.4.1 Switches reports13.4.2 Switch Port Data Rate Performance13.5 Case studies13.5.1 Server performance problem13.5.2 Disk performance problem in a Storwize V7000 subsystem13.5.3 Top volumes response time and I/O rate performance report13.5.4 Performance constraint alerts for SAN Volume Controller and Storwize V700013.5.5 Monitoring and diagnosing performance problems for a fabric13.5.6 Verifying the SAN Volume Controller and Fabric configuration by using Topology Viewer13.6 Monitoring in real time by using the SAN Volume Controller or Storwize V7000 GUI13.7 Manually gathering SAN Volume Controller statistics
14.1 Automating SAN Volume Controller and SAN environment documentation14.1.1 Naming conventions14.1.2 SAN fabrics documentation14.1.3 SAN Volume Controller14.1.4 Storage14.1.5 Technical Support information14.1.6 Tracking incident and change tickets14.1.7 Automated support data collection14.1.8 Subscribing to SAN Volume Controller support14.2 Storage management IDs14.3 Standard operating procedures14.3.1 Allocating and deallocating volumes to hosts14.3.2 Adding and removing hosts in SAN Volume Controller14.4 SAN Volume Controller code upgrade14.4.1 Preparing for the upgrade14.4.2 SAN Volume Controller upgrade from V5.1 to V6.214.4.3 Upgrading SVC clusters that are participating in Metro Mirror or Global Mirror14.4.4 SAN Volume Controller upgrade14.5 SAN modifications14.5.1 Cross-referencing HBA WWPNs14.5.2 Cross-referencing LUN IDs14.5.3 HBA replacement14.6 Hardware upgrades for SAN Volume Controller14.6.1 Adding SVC nodes to an existing cluster14.6.2 Upgrading SVC nodes in an existing cluster14.6.3 Moving to a new SVC cluster14.7 More information
15.1 Common problems15.1.1 Host problems15.1.2 SAN Volume Controller problems15.1.3 SAN problems15.1.4 Storage subsystem problems15.2 Collecting data and isolating the problem15.2.1 Host data collection15.2.2 SAN Volume Controller data collection15.2.3 SAN data collection15.2.4 Storage subsystem data collection15.3 Recovering from problems15.3.1 Solving host problems15.3.2 Solving SAN Volume Controller problems15.3.3 Solving SAN problems15.3.4 Solving back-end storage problems15.4 Mapping physical LBAs to volume extents15.4.1 Investigating a medium error by using lsvdisklba15.4.2 Investigating thin-provisioned volume allocation by using lsmdisklba15.5 Medium error logging15.5.1 Host-encountered media errors15.5.2 SAN Volume Controller-encountered medium errors
16.1 SAN Volume Controller upgrade with CF8 nodes and internal solid-state drives16.2 Moving an AIX server to another LPAR16.3 Migrating to new SAN Volume Controller by using Copy Services16.4 SAN Volume Controller scripting16.4.1 Connecting to the SAN Volume Controller by using a predefined SSH connection16.4.2 Scripting toolkit
IBM Redbooks publicationsOther resourcesReferenced websitesHelp from IBM

Overview

This IBM® Redbooks® publication captures several of the best practices based on field experience and describes the performance gains that can be achieved by implementing the IBM System Storage® SAN Volume Controller V6.2.

This book begins with a look at the latest developments with SAN Volume Controller V6.2 and reviews the changes in the previous versions of the product. It highlights configuration guidelines and best practices for the storage area network (SAN) topology, clustered system, back-end storage, storage pools and managed disks, volumes, remote copy services, and hosts. Then, this book provides performance guidelines for SAN Volume Controller, back-end storage, and applications. It explains how you can optimize disk performance with the IBM System Storage Easy Tier® function. Next, it provides best practices for monitoring, maintaining, and troubleshooting SAN Volume Controller. Finally, this book highlights several scenarios that demonstrate the best practices and performance guidelines.

This book is intended for experienced storage, SAN, and SAN Volume Controller administrators and technicians. Before reading this book, you must have advanced knowledge of the SAN Volume Controller and SAN environment. For background information, read the following Redbooks publications:

Implementing the IBM System Storage SAN Volume Controller V5.1, SG24-6423
Introduction to Storage Area Networks, SG24-5470

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Start your free trial

IBM System Storage SAN Volume Controller and Storwize V7000 Best Practices and Performance Guidelines

Jon Tate, Pawel Brodacki, Tilak Buneti, Christian Burns, Jana Jamsek, Erez Kirson, Marcin Tabinowski, Bosmat Tuv-El

IBM System Storage SAN Volume Controller and Storwize V7000 Best Practices and Performance Guidelines

Jon Tate, Angelo Bernasconi, Tiago Moreira Candelaria Bastos, Giulio Fiscella, Bruno Garcia Galle, Jana Jamsek, Antonio Rainero

IBM System Storage SAN Volume Controller, IBM Storwize V7000, and IBM FlashSystem 7200 Best Practices and Performance Guidelines

Jon Tate, Tiago Moreira Candelaria Bastos, Jon Herd, Sang-hyun Kim, Sergey Kubin, Dirk Peitzmann, Antonio Rainero

IBM SAN Volume Controller Best Practices and Performance Guidelines for IBM Spectrum Virtualize Version 8.4.2

Antonio Rainero, Carlton Beatty, David Green, Hartmut Lonzer, Jonathan Wilkie, Kendall Williams, Konrad Trojok, Mandy Stevens, Nezih Boyacioglu, Nils Olsson, Renato Santos, Rene Oehme, Sergey Kubin, Thales Noivo Ferreira, Uwe Schreiber, Vasfi Gucer

Publisher Resources

ISBN: 0738437115Purchase book

Overview

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,and much more.

You might also like

IBM System Storage SAN Volume Controller and Storwize V7000 Best Practices and Performance Guidelines

IBM System Storage SAN Volume Controller and Storwize V7000 Best Practices and Performance Guidelines

IBM System Storage SAN Volume Controller, IBM Storwize V7000, and IBM FlashSystem 7200 Best Practices and Performance Guidelines

IBM SAN Volume Controller Best Practices and Performance Guidelines for IBM Spectrum Virtualize Version 8.4.2

Publisher Resources

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.