You are previewing IBM SAN Solution Design Best Practices for VMware vSphere ESXi.
O'Reilly logo
IBM SAN Solution Design Best Practices for VMware vSphere ESXi

Book Description

In this IBM® Redbooks® publication, we describe recommendations based on an IBM b-type storage area network (SAN) environment that is utilizing VMware vSphere ESXi. We describe the hardware and software and the unique features that they bring to the marketplace. We then highlight those features and how they apply to the SAN environment, and the best practices for ensuring that you get the best out of your SAN.

For background reading, we recommend the following Redbooks publications:
- Introduction to Storage Area Networks and System Networking, SG24-5470
- IBM System Storage SAN Volume Controller Best Practices and Performance Guidelines, SG24-7521
- IBM System Storage SAN Volume Controller and Storwize V7000 Replication Family Services, SG24-7574
- Implementing the IBM System Storage SAN Volume Controller V6.3, SG24-7933
- IBM SAN Volume Controller Stretched Cluster with PowerVM and PowerHA, SG24-8142
- Implementing the IBM SAN Volume Controller and FlashSystem 820, SG24-8172
- IBM System Storage DS8000 Copy Services for Open Systems, SG24-6788
- IBM System Storage DS8000: Host Attachment and Interoperability, SG24-8887

This book is aimed at pre- and post-sales support, system administrators, and storage administrators.

Table of Contents

  1. Front cover
  2. Notices
    1. Trademarks
  3. Preface
    1. Authors
    2. Now you can become a published author, too!
    3. Comments welcome
    4. Stay connected to IBM Redbooks
  4. Chapter 1. Introduction
    1. 1.1 VMware vSphere ESXi
    2. 1.2 IBM Disk Storage Systems
      1. 1.2.1 IBM Storwize V3700
      2. 1.2.2 IBM Storwize V7000
      3. 1.2.3 IBM SAN Volume Controller
    3. 1.3 IBM System Networking SAN b-type family
      1. 1.3.1 IBM System Storage SAN24B-5 (2498-F24)
      2. 1.3.2 IBM System Storage SAN48B-5 (2498-F48)
      3. 1.3.3 IBM System Storage SAN96B-5 (2498-F96 and 2498-N96)
      4. 1.3.4 IBM System Storage SAN384B-2 (2499-416) and SAN768B-2 (2499-816)
  5. Chapter 2. General SAN design and best practices
    1. 2.1 16 Gbps Fibre Channel
      1. 2.1.1 Overview of 16 Gbps Fibre Channel
    2. 2.2 Physical patching
      1. 2.2.1 Using a structured approach
      2. 2.2.2 Modular data cabling
      3. 2.2.3 Cabling high-density, high port-count fiber equipment
      4. 2.2.4 Using color to identify cables
      5. 2.2.5 Establishing a naming scheme
      6. 2.2.6 Patch cables
      7. 2.2.7 Patch panels
      8. 2.2.8 Horizontal and backbone cables
      9. 2.2.9 Horizontal cable managers
      10. 2.2.10 Vertical cable managers
      11. 2.2.11 Overhead cable pathways
      12. 2.2.12 Cable ties
      13. 2.2.13 Implementing the cabling infrastructure
      14. 2.2.14 Testing the links
      15. 2.2.15 Building a common framework for the racks
      16. 2.2.16 Preserving the infrastructure
      17. 2.2.17 Documentation
      18. 2.2.18 Stocking spare cables
      19. 2.2.19 Best practices for managing the cabling
      20. 2.2.20 Summary
    3. 2.3 Switch interconnections
      1. 2.3.1 Inter-switch link
      2. 2.3.2 Inter-chassis links
      3. 2.3.3 Fabric shortest path first
    4. 2.4 Device placement
      1. 2.4.1 Traffic locality
      2. 2.4.2 Fan-in ratios and oversubscription
    5. 2.5 Data flow considerations
      1. 2.5.1 Congestion in the fabric
      2. 2.5.2 Traffic versus frame congestion
      3. 2.5.3 Sources of congestion
      4. 2.5.4 Mitigating congestion with edge hold time
    6. 2.6 Host bus adapter
      1. 2.6.1 b-type host bus adapters
      2. 2.6.2 QLogic
      3. 2.6.3 Emulex
    7. 2.7 Zoning
    8. 2.8 Redundancy and resiliency
      1. 2.8.1 Single point of failure
    9. 2.9 Topologies
      1. 2.9.1 Core-edge topology
      2. 2.9.2 Edge-core-edge topology
      3. 2.9.3 Full-mesh topology
    10. 2.10 Distance
      1. 2.10.1 Buffer allocation
      2. 2.10.2 Fabric interconnectivity over Fibre Channel at longer distances
      3. 2.10.3 Fibre Channel over IP
      4. 2.10.4 FCIP with FCR
      5. 2.10.5 Using EX_Ports and VEX_Ports
      6. 2.10.6 Advanced FCIP configuration
      7. 2.10.7 FCIP design best practices
      8. 2.10.8 FCIP Trunking
      9. 2.10.9 Virtual Fabrics
      10. 2.10.10 Ethernet Interface Sharing
      11. 2.10.11 Workloads
      12. 2.10.12 Intel based virtualization storage access
    11. 2.11 Security
      1. 2.11.1 Zone management: Dynamic Fabric Provisioning
      2. 2.11.2 Zone management: Duplicate WWNs
      3. 2.11.3 Role-based access controls
      4. 2.11.4 Access control lists
      5. 2.11.5 Policy database distribution
      6. 2.11.6 In-flight encryption and compression: b-type (16 Gbps) platforms only
      7. 2.11.7 In-flight encryption and compression guidelines
    12. 2.12 Monitoring
      1. 2.12.1 Fabric Watch
      2. 2.12.2 RAS log
      3. 2.12.3 Audit log
      4. 2.12.4 SAN Health
      5. 2.12.5 Design guidelines
      6. 2.12.6 Monitoring and notifications
    13. 2.13 Scalability, supportability, and performance
  6. Chapter 3. General practices for VMware
    1. 3.1 VMware Pluggable Storage Architecture
      1. 3.1.1 VMware NMP Flow of I/O
      2. 3.1.2 Path Selection Plug-ins
    2. 3.2 Asymmetric Logical Unit Access
      1. 3.2.1 Path trashing
      2. 3.2.2 Finding the optimized paths
    3. 3.3 VMware vStorage APIs for Storage Awareness
      1. 3.3.1 Profile Driven Storage
    4. 3.4 Storage I/O Control
      1. 3.4.1 SIOC limitations and requirements
      2. 3.4.2 Storage I/O Control congestion latency
      3. 3.4.3 Conclusion
    5. 3.5 Storage Distributed Resource Scheduler
      1. 3.5.1 Migration recommendations
      2. 3.5.2 SDRS I/O balancing
      3. 3.5.3 SDRS space balancing
      4. 3.5.4 Schedule SDRS for off-peak hours
      5. 3.5.5 Conclusion
    6. 3.6 Virtual Machine File System
      1. 3.6.1 VMFS extents
      2. 3.6.2 Disk alignment
      3. 3.6.3 Virtual machine files
    7. 3.7 VMware vMotion
    8. 3.8 VMware storage vMotion
    9. 3.9 vStorage APIs for Array Integration
      1. 3.9.1 Requirements
      2. 3.9.2 Confirming VAAI Hardware Acceleration is detected
      3. 3.9.3 Setting the data transfer chunk size
    10. 3.10 Raw Device Mapping
    11. 3.11 VMware thin provisioning
      1. 3.11.1 Using VMFS thin provisioning
      2. 3.11.2 Thin provisioning prerequisites
      3. 3.11.3 Thin provisioning general guidelines
      4. 3.11.4 Thin on thin?
    12. 3.12 IBM Storage Management Console for VMware vCenter
      1. 3.12.1 Conclusion
    13. 3.13 General recommendation
  7. Chapter 4. General practices for storage
    1. 4.1 Configuring and servicing external storage systems
      1. 4.1.1 General guidelines for SAN Volume Controller
      2. 4.1.2 General Guidelines for Storwize V7000
      3. 4.1.3 Configuring the Storwize V3700
      4. 4.1.4 Storwize family presets
    2. 4.2 Disk
      1. 4.2.1 Input/output operations per second
      2. 4.2.2 Disk types
    3. 4.3 MDisks and volumes
      1. 4.3.1 Disk tiering
      2. 4.3.2 Logical disk configuration guidelines for storage systems
      3. 4.3.3 RAID configuration guidelines for storage systems
      4. 4.3.4 Optimal storage pool configuration guidelines for storage systems
      5. 4.3.5 FlashCopy mapping guidelines for storage systems
      6. 4.3.6 Image mode volumes and data migration guidelines for storage systems
      7. 4.3.7 Configuring a balanced storage system
    4. 4.4 Volumes
      1. 4.4.1 Thin provisioning
    5. 4.5 Back-end
    6. 4.6 Front-end/fabric
      1. 4.6.1 4-way multipathing
      2. 4.6.2 8-way multipathing
      3. 4.6.3 Greater than 8-way multipathing
    7. 4.7 VMware considerations
      1. 4.7.1 Configuring the QLogic HBA for hosts running the VMware OS
      2. 4.7.2 Queue depth
    8. 4.8 Maintenance considerations
    9. 4.9 Putting it all together
    10. 4.10 References
  8. Chapter 5. Business continuity and disaster recovery
    1. 5.1 Continuity and recovery solutions
    2. 5.2 IBM Replication Family Services
      1. 5.2.1 FlashCopy
      2. 5.2.2 Metro Mirror
      3. 5.2.3 Global Mirror
      4. 5.2.4 Image mode migration and volume mirroring migration
    3. 5.3 IBM SAN Volume Controller Stretched Cluster
    4. 5.4 VMware vSphere Fault Tolerance
    5. 5.5 VMware vSphere vMotion and VMware vSphere HA
      1. 5.5.1 VMware vSphere vMotion
      2. 5.5.2 VMware vSphere High Availability
      3. 5.5.3 VMware vSphere Metro Storage Cluster
      4. 5.5.4 VMware vCenter Site Recovery Manager
    6. 5.6 Storage Replication Adapter for IBM SAN Volume Controller
    7. 5.7 Backup and restore solutions
    8. 5.8 Traditional backup and restore with Tivoli Storage Manager
      1. 5.8.1 Disaster recovery manager
    9. 5.9 Tivoli Storage Manager for Virtual Environments
    10. 5.10 VMware vSphere Data Protection
  9. Chapter 6. Entry level scenario
    1. 6.1 Storage area network
      1. 6.1.1 Topology
      2. 6.1.2 Naming convention and zoning scheme
      3. 6.1.3 16 Gb Fibre Channel host bus adapters
    2. 6.2 Storage subsystem
      1. 6.2.1 Preparing for Fibre Channel attachment
      2. 6.2.2 VMware ESXi installation
      3. 6.2.3 VMware ESXi multipathing
      4. 6.2.4 VMware license considerations
      5. 6.2.5 VMware support contracts
      6. 6.2.6 Backup and disaster recovery
  10. Chapter 7. Midrange level scenario
    1. 7.1 Storage area network
      1. 7.1.1 Topology
      2. 7.1.2 Naming convention and zoning scheme
    2. 7.2 Storage subsystem
      1. 7.2.1 Preparing for Fibre Channel attachment
      2. 7.2.2 VMware ESXi installation
      3. 7.2.3 VMware ESXi multipathing
    3. 7.3 VMware considerations
      1. 7.3.1 Backup and disaster recovery
  11. Chapter 8. Enterprise scenario
    1. 8.1 Introduction
    2. 8.2 Fabric types
      1. 8.2.1 Edge-core-edge topology
      2. 8.2.2 Device placement
    3. 8.3 Edge-core-edge design
      1. 8.3.1 Storage edge
      2. 8.3.2 Trunk groups
      3. 8.3.3 Storwize V7000
      4. 8.3.4 SAN Volume Controller and core
      5. 8.3.5 Host edge
    4. 8.4 Zoning
      1. 8.4.1 Types of zoning
      2. 8.4.2 Prezoning tips and shortcuts
      3. 8.4.3 SAN Volume Controller internode communications zone
      4. 8.4.4 SAN Volume Controller storage zones
      5. 8.4.5 Storwize V7000 storage subsystem
      6. 8.4.6 SAN Volume Controller host zones
      7. 8.4.7 Standard SAN Volume Controller zoning configuration
      8. 8.4.8 Aliases
      9. 8.4.9 Zones
      10. 8.4.10 Zoning with multiple SAN Volume Controller clustered systems
      11. 8.4.11 Split storage subsystem configurations
    5. 8.5 Switch domain IDs
    6. 8.6 Tying it all together
      1. 8.6.1 Setting up aliases
    7. 8.7 VMware enterprise level
      1. 8.7.1 Backup and disaster recovery
    8. 8.8 References
  12. Related publications
    1. IBM Redbooks
    2. Help from IBM
  13. Back cover