InfoScale™ 9.0 Storage Foundation and High Availability Solutions Solutions Guide - Windows

Last Published:
Product(s): InfoScale & Storage Foundation (9.0)
Platform: Windows
  1. Section I. Introduction
    1. Introducing Storage Foundation and High Availability Solutions
      1.  
        About Quick Recovery
      2.  
        About high availability
      3.  
        About campus clusters
      4.  
        About disaster recovery
      5.  
        About Microsoft clustering solutions
      6.  
        How this guide is organized
    2. Using the Solutions Configuration Center
      1.  
        About the Solutions Configuration Center
      2.  
        Starting the Solutions Configuration Center
      3.  
        Options in the Solutions Configuration Center
      4.  
        About launching wizards from the Solutions Configuration Center
      5.  
        Remote and local access to Solutions wizards
      6.  
        Solutions wizards and logs
      7.  
        Workflows in the Solutions Configuration Center
    3. SFW best practices for storage
      1. Best practices for storage availability
        1.  
          Adding software mirrors for critical data
        2.  
          Locating data objects for optimum recovery
        3.  
          Managing three-way software mirrors for reliability
        4.  
          Software striping and mirroring on top of hardware RAID for high availability
      2. Best practices configuring SFW disk groups for availability
        1.  
          Configuring disk groups for separate storage capacity pools or common pools
        2.  
          Allocating disk groups for availability in clusters
      3. Best practices for storage performance
        1.  
          Host-based mirroring for increased read performance and failure tolerance
        2.  
          Software striping across hardware for increased performance
        3.  
          Using software RAID 5 for read-mostly data
      4. Best practices for I/O performance tuning
        1.  
          Striping for I/O-request-intensive applications
        2.  
          Striping for data-transfer-intensive applications
      5. Best practices for storage capacity management
        1.  
          Managing storage allocation for flexibility
        2.  
          Aggregating hardware RAID for very large volumes
        3.  
          Managing unallocated space for free space savings
        4.  
          Reserving spares for failure-tolerant volume recovery
  2. Section II. Quick Recovery
    1. Quick Recovery overview
      1.  
        About the Quick Recovery solution
      2.  
        Need for implementing the SFW Quick Recovery solution
      3.  
        Understanding the underlying components of SFW's Quick Recovery process
      4. Overview of the Quick Recovery process
        1.  
          Creating initial snapshots
        2.  
          Refreshing a snapshot
        3.  
          Recovering a database
      5. Other applications for point-in-time snapshots
        1.  
          Off-host backups
        2.  
          Reporting and analysis
        3.  
          Application testing and training
    2. Quick Recovery example
      1. Example of quick recovery of an Oracle database
        1.  
          Creating a split-mirror snapshot of the database
        2.  
          Recovering the database using the split-mirror snapshot and database logs
      2.  
        Tips and references about FlashSnap
  3. Section III. High Availability
    1. High availability: Overview
      1.  
        About high availability
      2.  
        About clusters
      3. How VCS monitors storage components
        1.  
          Shared storage - if you use NetApp filers
        2.  
          Shared storage - if you use SFW to manage cluster dynamic disk groups
        3.  
          Shared storage - if you use Windows LDM to manage shared disks
        4.  
          Non-shared storage - if you use SFW to manage dynamic disk groups
        5.  
          Non-shared storage - if you use Windows LDM to manage local disks
        6.  
          Non-shared storage - if you use VMware storage
    2. Deploying InfoScale Enterprise for high availability: New installation
      1.  
        About the high availability solution
      2.  
        Tasks for a new high availability (HA) installation - additional applications
      3.  
        Reviewing the InfoScale installation requirements
      4. Notes and recommendations for cluster and application configuration
        1.  
          IPv6 support
      5.  
        Reviewing the configuration
      6.  
        Configuring the storage hardware and network
      7.  
        About installing the Arctera InfoScale™ products
      8. Configuring disk groups and volumes
        1. Planning disk groups and volumes
          1.  
            Planning your File Share storage
          2.  
            Planning your IIS storage
          3.  
            Planning your storage for additional applications
          4.  
            Considerations for a fast failover configuration
        2.  
          Creating a dynamic disk group
        3.  
          Creating dynamic volumes
        4.  
          About managing disk groups and volumes
        5.  
          Importing a disk group and mounting a volume
        6.  
          Unmounting a volume and deporting a disk group
      9. Configuring the cluster using the Cluster Configuration Wizard
        1.  
          Configuring notification
      10. About modifying the cluster configuration
        1.  
          Adding nodes to a cluster
        2.  
          Removing nodes from a cluster
        3.  
          Reconfiguring a cluster
        4. Configuring the ClusterService group
          1.  
            Configuring notification
          2.  
            Configuring the wide-area connector process for global clusters
        5.  
          Deleting a cluster configuration
      11. About installing and configuring the application or server role
        1.  
          About configuring a File Share server role
        2.  
          About installing and configuring the IIS application
        3.  
          About installing additional applications
      12. Configuring the service group
        1. About configuring file shares
          1.  
            Before you configure a file share service group
          2.  
            Configuring file shares using the wizard
          3.  
            Creating non-scoped file shares configured with VCS
          4.  
            Making non-scoped file shares accessible while using virtual server name or IP address if NetBIOS and WINS are disabled
        2. About configuring IIS sites
          1.  
            Before you configure an IIS service group
          2.  
            Fixing the IPv6 address configuration for FTP sites
          3.  
            Installing IIS on Windows Server Core
          4.  
            Configuring an IIS service group using the wizard
        3. About configuring applications using the Application Configuration Wizard
          1.  
            Before you configure service groups using the Application Configuration wizard
          2. Adding resources to a service group
            1.  
              Configuring a GenericService resource
            2.  
              Configuring processes
            3.  
              Configuring a ServiceMonitor resource
            4.  
              Configuring VCS components
          3.  
            Configuring service groups using the Application Configuration Wizard
        4. About configuring the Oracle service group using the wizard
          1.  
            Prerequisites for configuring the Oracle service group
          2.  
            Creating an Oracle service group using the wizard
          3.  
            Configuring dependent services
        5.  
          Enabling fast failover for disk groups (optional)
      13.  
        Configuring the service group in a non-shared storage environment
      14.  
        Verifying the cluster configuration
      15.  
        Possible tasks after completing the configuration
      16.  
        Adding nodes to a cluster
      17. Modifying the application service groups
        1.  
          Modifying a file share service group using the wizard
        2.  
          Modifying an IIS service group using the wizard
        3.  
          Modifying an application service group
        4. About modifying an Oracle service group
          1.  
            Prerequisites for modifying the Oracle service group
          2.  
            Modifying an Oracle service group
          3.  
            Deleting an Oracle service group
          4. Configuring detail monitoring
            1.  
              Enabling detail monitoring
            2.  
              Disabling detail monitoring
    3. Adding DMP to a clustering configuration
      1.  
        About Dynamic Multi-Pathing
      2.  
        Overview of configuration tasks for adding DMP DSMs
      3.  
        Reviewing the prerequisites
      4.  
        Reviewing the configuration
      5.  
        Setting up DMP in a new cluster configuration
      6.  
        Setting up DMP in an existing cluster configuration
  4. Section IV. Campus Clustering
    1. Introduction to campus clustering
      1.  
        About Campus Clusters
      2.  
        Sample campus cluster configuration
      3.  
        Differences between campus clusters and local clusters
    2. Deploying InfoScale Enterprise for campus cluster
      1.  
        About the Campus Cluster solution
      2. Notes and recommendations for cluster and application configuration
        1.  
          IPv6 support
        2.  
          Campus cluster requirements
      3. Reviewing the configuration
        1.  
          Overview of campus clustering with VCS
        2.  
          Reinstating faulted hardware
        3.  
          Setting the ForceImport attribute
      4.  
        Installing and configuring the hardware
      5.  
        Configuring the storage hardware and network
      6.  
        About installing the Arctera InfoScale™ products
      7. Configuring the cluster using the Cluster Configuration Wizard
        1.  
          Configuring notification
      8. Creating disk groups and volumes
        1.  
          About cluster disk groups and volumes
        2.  
          Example disk group and volume configuration in campus cluster
        3.  
          Considerations when creating disks and volumes for campus clusters
        4.  
          Viewing the available disk storage
        5.  
          Creating a dynamic disk group
        6.  
          Adding disks to campus cluster sites
        7.  
          Creating volumes for campus clusters
      9. Installing the application on cluster nodes
        1.  
          About configuring a File Share server role in a campus cluster
        2.  
          About installing and configuring the IIS application in a campus cluster
        3.  
          About installing additional applications in a campus cluster
        4.  
          Deporting and importing a disk group in a campus cluster
      10.  
        Configuring service groups
      11.  
        Verifying the cluster configuration
  5. Section V. Replicated Data Clusters
    1. Introduction to Replicated Data Clusters
      1.  
        About Replicated Data Clusters
      2.  
        How VCS Replicated Data Clusters work
      3. Setting up a Replicated Data Cluster configuration
        1.  
          Setting up replication
        2.  
          Configuring the service groups
      4.  
        Migrating the service group
    2. Deploying Replicated Data Clusters: New application installation
      1.  
        Tasks for a new replicated data cluster installation - additional applications
      2. Notes and recommendations for cluster and application configuration
        1.  
          IPv6 support
      3.  
        Sample configuration
      4.  
        Configuring the storage hardware and network
      5.  
        About installing the Arctera InfoScale™ products
      6.  
        Setting up security for Volume Replicator
      7. Configuring the cluster using the Cluster Configuration Wizard
        1.  
          Configuring notification
      8. Configuring disk groups and volumes
        1. Planning disk groups and volumes
          1.  
            Planning your File Share storage
          2.  
            Planning your IIS storage
          3.  
            Planning your storage for additional applications
        2.  
          Creating a dynamic disk group
        3.  
          Creating dynamic volumes
        4. About managing disk groups and volumes
          1.  
            Importing a disk group and mounting a volume
          2.  
            Unmounting a volume and deporting a disk group
      9. Installing and configuring the application or server role
        1.  
          Configuring a File Share server role
        2.  
          Installing and configuring the IIS application
        3.  
          Installing additional applications
      10. Configuring the service group
        1. About configuring file shares
          1.  
            Before you configure a file share service group
          2.  
            Configuring file shares using the wizard
          3.  
            Creating non-scoped file shares configured with VCS
          4.  
            Making non-scoped file shares accessible while using virtual server name or IP address if NetBIOS and WINS are disabled
        2. About configuring IIS sites
          1.  
            Before you configure an IIS service group
          2.  
            Fixing the IPv6 address configuration for FTP sites
          3.  
            Installing IIS on Windows Server Core
          4.  
            Configuring an IIS service group using the wizard
        3. About configuring applications using the Application Configuration Wizard
          1.  
            Before you configure service groups using the Application Configuration wizard
          2. Adding resources to a service group
            1.  
              Configuring a GenericService resource
            2.  
              Configuring processes
            3.  
              Configuring a ServiceMonitor resource
            4.  
              Configuring VCS components
          3.  
            Configuring service groups using the Application Configuration Wizard
      11.  
        Creating the primary system zone for the application service group
      12.  
        Verifying the cluster configuration
      13.  
        Creating a parallel environment in the secondary zone
      14.  
        Adding nodes to a cluster
      15.  
        Creating the Replicated Data Sets with the wizard
      16. Configuring a RVG service group for replication
        1.  
          Creating the RVG service group
        2. Configuring the resources in the RVG service group for RDC replication
          1.  
            Configuring the IP and NIC resources
          2. Configuring the VMDg or VMNSDg resources for the disk groups
            1.  
              Modifying the DGGuid attribute for the new disk group resource in the RVG service group
          3.  
            Adding the Volume Replicator RVG resources for the disk groups
          4.  
            Linking the Volume Replicator RVG resources to establish dependencies
          5.  
            Deleting the VMDg or VMNSDg resource from the application service group
        3. Configuring the RVG Primary resources
          1.  
            Creating the RVG Primary resources
          2.  
            Linking the RVG Primary resources to establish dependencies
          3.  
            Bringing the RVG Primary resources online
        4.  
          Configuring the primary system zone for the RVG service group
      17.  
        Setting a dependency between the service groups
      18. Adding the nodes from the secondary zone to the RDC
        1.  
          Adding the nodes from the secondary zone to the RVG service group
        2.  
          Configuring secondary zone nodes in the RVG service group
        3.  
          Configuring the RVG service group NIC resource for fail over (VMNSDg only)
        4.  
          Configuring the RVG service group IP resource for failover
        5.  
          Configuring the RVG service group VMNSDg resources for fail over
        6.  
          Adding the nodes from the secondary zone to the application service group
        7.  
          Configuring the zones in the application service group
        8.  
          Configuring the application service group IP resource for fail over (VMNSDg only)
        9.  
          Configuring the application service group NIC resource for fail over (VMNSDg only)
      19. Verifying the RDC configuration
        1.  
          Bringing the service group online
        2.  
          Switching online nodes
      20.  
        Additional instructions for GCO disaster recovery
  6. Section VI. Disaster Recovery
    1. Disaster recovery: Overview
      1.  
        About a disaster recovery solution
      2.  
        Need for implementing a disaster recovery solution
      3.  
        Overview of the recovery process
      4. Components of Volume Replicator that enable disaster recovery
        1.  
          Understanding replication
        2. Modes of replication
          1.  
            Synchronous replication
          2.  
            Asynchronous replication
          3.  
            Synchronous override replication
        3.  
          Features of Volume Replicator that help in disaster recovery
    2. Deploying disaster recovery: New application installation
      1.  
        Tasks for a new disaster recovery installation - additional applications
      2.  
        Tasks for setting up DR in a non-shared storage environment
      3. Notes and recommendations for cluster and application configuration
        1.  
          IPv6 support
      4. Reviewing the configuration
        1.  
          Supported disaster recovery configurations for service group dependencies
      5.  
        Configuring the storage hardware and network
      6. About managing disk groups and volumes
        1.  
          Importing a disk group and mounting a volume
        2.  
          Unmounting a volume and deporting a disk group
      7. Setting up the secondary site: Configuring SFW HA and setting up a cluster
        1. About installing the Arctera InfoScale™ products
          1. Installing the server components using the installation wizard
            1.  
              Applying the selected installation and product options to multiple systems
          2.  
            Installing the client components
        2.  
          Configuring the cluster using the Cluster Configuration Wizard
        3.  
          Configuring notification
      8.  
        Verifying that your application or server role is configured for HA at the primary site
      9. Setting up your replication environment
        1.  
          Setting up security for Volume Replicator
        2. Requirements for EMC SRDF array-based hardware replication
          1.  
            Software requirements for configuring EMC SRDF
          2.  
            Replication requirements for EMC SRDF
        3. Requirements for Hitachi TrueCopy array-based hardware replication
          1.  
            Software requirements for Hitachi TrueCopy
          2.  
            Replication requirements for Hitachi TrueCopy
      10.  
        Assigning user privileges (secure clusters only)
      11. About configuring disaster recovery with the DR wizard
        1.  
          Configuring disaster recovery with the DR wizard
      12.  
        Cloning the storage on the secondary site using the DR wizard (Volume Replicator replication option)
      13.  
        Creating temporary storage on the secondary site using the DR wizard (array-based replication)
      14. Installing and configuring the application or server role (secondary site)
        1.  
          Installing the FileShare application
        2.  
          Installing the IIS application
        3.  
          Installing additional applications
      15.  
        Cloning the service group configuration from the primary site to the secondary site
      16.  
        Configuring the application service group in a non-shared storage environment
      17. Configuring replication and global clustering
        1.  
          Configuring Volume Replicator replication and global clustering
        2. Configuring EMC SRDF replication and global clustering
          1.  
            Optional settings for EMC SRDF
        3. Configuring Hitachi TrueCopy replication and global clustering
          1.  
            Optional settings for HTC
        4.  
          Configuring global clustering only
      18.  
        Creating the replicated data sets (RDS) for Volume Replicator replication
      19.  
        Creating the Volume Replicator RVG service group for replication
      20. Configuring the global cluster option for wide-area failover
        1.  
          Linking clusters: Adding a remote cluster to a local cluster
        2.  
          Converting a local service group to a global service group
        3.  
          Bringing a global service group online
      21.  
        Verifying the disaster recovery configuration
      22.  
        Adding multiple DR sites (optional)
      23. Possible task after creating the DR environment: Adding a new failover node to a Volume Replicator environment
        1.  
          Preparing the new node
        2.  
          Preparing the existing DR environment
        3.  
          Modifying the replication and application service groups
        4.  
          Reversing replication direction
      24. Maintaining: Normal operations and recovery procedures (Volume Replicator environment)
        1.  
          Normal operations: Monitoring the status of the replication
        2.  
          Performing planned migration
        3. Disaster recovery procedures
          1.  
            Restoring the primary host
      25.  
        Recovery procedures for service group dependencies
    3. Testing fault readiness by running a fire drill
      1.  
        About disaster recovery fire drills
      2. About the Fire Drill Wizard
        1.  
          About Fire Drill Wizard general operations
        2. About Fire Drill Wizard operations in a Volume Replicator environment
          1.  
            Preparing the fire drill configuration
          2.  
            About running the fire drill
          3.  
            About restoring the fire drill configuration
          4.  
            About deleting the fire drill configuration
        3.  
          About Fire Drill Wizard operations in a Hitachi TrueCopy or EMC SRDF environment
      3.  
        About post-fire drill scripts
      4.  
        Tasks for configuring and running fire drills
      5. Prerequisites for a fire drill
        1.  
          Prerequisites for a fire drill in a Volume Replicator environment
        2.  
          Prerequisites for a fire drill in a Hitachi TrueCopy environment
        3.  
          Prerequisites for a fire drill in an EMC SRDF environment
      6. Preparing the fire drill configuration
        1.  
          System Selection panel details
        2.  
          Service Group Selection panel details
        3.  
          Secondary System Selection panel details
        4.  
          Fire Drill Service Group Settings panel details
        5.  
          Disk Selection panel details
        6.  
          Hitachi TrueCopy Path Information panel details
        7.  
          HTCSnap Resource Configuration panel details
        8.  
          SRDFSnap Resource Configuration panel details
        9.  
          Fire Drill Preparation panel details
      7.  
        Running a fire drill
      8.  
        Re-creating a fire drill configuration that has changed
      9.  
        Restoring the fire drill system to a prepared state
      10. Deleting the fire drill configuration
        1.  
          Fire Drill Deletion panel details
      11.  
        Considerations for switching over fire drill service groups
  7. Section VII. Microsoft Clustering Solutions
    1. Microsoft clustering solutions overview
      1.  
        About Microsoft clustering with high availability
      2.  
        About Microsoft clustering with Volume Replicator
      3.  
        About Microsoft clustering with campus clustering
      4. About the SFW-Microsoft clustering-Volume Replicator configuration
        1.  
          Configuring the quorum device for high availability
    2. Deploying SFW with Microsoft failover clustering
      1.  
        Tasks for deploying InfoScale Storage with Microsoft failover clustering
      2.  
        Reviewing the configuration
      3.  
        Configuring the storage hardware and network
      4.  
        Establishing a Microsoft failover cluster
      5. Tasks for installing InfoScale Foundation or InfoScale Storage for Microsoft failover clustering
        1.  
          Pre-installation task: moving the online groups
        2.  
          About installing the Arctera InfoScale™ products
        3. Installing the server components using the installation wizard
          1.  
            Applying the selected installation and product options to multiple systems
          2.  
            Registering the InfoScale Storage resource DLLs
        4.  
          Installing the client components
        5.  
          Post-installation task: moving the online groups
      6. Creating SFW disk groups and volumes
        1.  
          Planning disk groups and volumes
        2.  
          Creating a dynamic disk group
        3.  
          Creating dynamic volumes
      7.  
        Creating a group for the application in the failover cluster
      8.  
        Installing the application on cluster nodes
      9.  
        Completing the setup of the application group in the failover cluster
      10. Implementing a dynamic quorum resource
        1.  
          Creating a dynamic cluster disk group and a mirrored volume for the quorum resource
        2.  
          Adding a VMDg resource for the quorum
        3.  
          Changing the quorum resource to a dynamic mirrored quorum resource
      11.  
        Verifying the cluster configuration
      12.  
        Configuring InfoScale Storage in an existing Microsoft Failover Cluster
    3. Deploying SFW with Microsoft failover clustering in a campus cluster
      1.  
        Tasks for deploying InfoScale Storage with Microsoft failover clustering in a campus cluster
      2. Reviewing the configuration
        1.  
          Overview of campus clustering with Microsoft clustering
        2. Campus cluster failure with Microsoft clustering scenarios
          1.  
            Dealing with a failover situation
        3. Microsoft clustering quorum and quorum arbitration
          1.  
            Quorum
          2.  
            Cluster ownership of the quorum resource
          3.  
            The vxclus utility
      3.  
        Configuring the storage hardware and network
      4. Establishing a Microsoft failover cluster
        1.  
          Connecting the two nodes
      5. Tasks for installing InfoScale Foundation or InfoScale Storage for Microsoft failover clustering
        1.  
          Pre-installation task: moving the online groups
        2.  
          About installing the Arctera InfoScale™ products
        3.  
          Post-installation task: moving the online groups
      6. Creating disk groups and volumes
        1.  
          Example disk group and volume configuration in campus cluster
        2.  
          Considerations when creating disks and volumes for campus clusters
        3.  
          Viewing the available disk storage
        4.  
          Creating a dynamic disk group
        5.  
          Adding disks to campus cluster sites
        6.  
          Creating volumes for campus clusters
      7. Implementing a dynamic quorum resource
        1.  
          Creating a dynamic cluster disk group and a mirrored volume for the quorum resource
        2.  
          Adding a VMDg resource for the quorum
        3.  
          Changing the quorum resource to a dynamic mirrored quorum resource
      8.  
        Setting up a group for the application in the failover cluster
      9. Installing the application on the cluster nodes
        1.  
          Pointers for installing the application on the first node
        2.  
          Pointers for installing the application on the second node
      10.  
        Completing the setup of the application group in the cluster
      11.  
        Verifying the cluster configuration
    4. Deploying SFW and VVR with Microsoft failover clustering
      1.  
        Tasks for deploying InfoScale Storage and Volume Replicator with Microsoft failover clustering
      2. Part 1: Setting up the cluster on the primary site
        1. Reviewing the prerequisites and the configuration
          1.  
            Reviewing the configuration
        2.  
          Installing and configuring the hardware
        3.  
          Installing Windows and configuring network settings
        4.  
          Establishing a Microsoft failover cluster
        5.  
          Installing InfoScale Storage (primary site)
        6.  
          Setting up security for Volume Replicator
        7.  
          Creating SFW disk groups and volumes
        8.  
          Completing the primary site configuration
      3. Part 2: Setting up the cluster on the secondary site
        1.  
          Repeating cluster configuration steps for the secondary site
      4. Part 3: Adding the Volume Replicator components for replication
        1.  
          Volume Replicator components overview
        2.  
          Configuring the Replicator Log volumes for Volume Replicator
        3.  
          Creating the Replicated Data Sets with the wizard
        4.  
          Creating resources for Volume Replicator
        5. Creating an RVG resource and setting the dependencies
          1.  
            Creating an RVG resource
          2.  
            Setting the application resource dependency on the RVG resource
      5. Part 4: Maintaining normal operations and recovery procedures
        1.  
          Normal operations: Monitoring the status of the replication
        2.  
          Performing planned migration
        3. Disaster recovery procedures
          1.  
            Bringing up the application on the secondary host
          2.  
            Restoring the primary host
  8. Section VIII. Server Consolidation
    1. Server consolidation overview
      1.  
        Server consolidation definition
      2.  
        Need for implementing server consolidation
      3.  
        Advantages of using SFW with server consolidation
      4.  
        Overview of the server consolidation process
    2. Server consolidation configurations
      1. Typical server consolidation configuration
        1.  
          Proof of concept
      2. Server consolidation configuration 1 - many to one
        1. About this configuration
          1.  
            Proof of concept
          2.  
            Phased approach: Flexible use of storage devices
        2.  
          Preparing to consolidate
        3.  
          Migrating the data to the large server
        4.  
          Adding the storage array
        5.  
          Completing the consolidation process
      3. Server consolidation configuration 2 - many to two: Adding clustering and DMP
        1. About this configuration
          1.  
            More on DMP paths
        2.  
          Adding the new hardware
        3.  
          Establishing the Microsoft failover cluster
        4.  
          Adding SFW support to the cluster
        5.  
          Setting up Microsoft failover cluster groups for the applications
        6.  
          Installing applications on the second computer
        7.  
          Completing the setup of the application group in the Microsoft cluster
        8.  
          Changing the quorum resource to the dynamic quorum resource
        9.  
          Verifying the cluster configuration
        10.  
          Enabling DMP
      4. SFW features that support server consolidation
        1.  
          Automatic volume growth
        2.  
          Features that support storage in a SAN
        3.  
          Performance monitoring

Striping for I/O-request-intensive applications

A good compromise stripe unit size for I/O-request-intensive applications is one that results in a 3% to 5% probability of splitting in a uniform distribution of requests. For example, a 2 KB (four-block) database page size would have an ideal stripe unit size of 100 blocks. This would typically be rounded up to the nearest power of two (128 blocks, or 65,536 bytes) for simplicity.

I/O-request-intensive applications are typically characterized by small (for example, 2 to 16 KB) data transfers for each request. These applications are I/O bound because they make so many I/O requests, not because they transfer large amounts of data.

For example, an application that makes 1,000 I/O requests per second with an average request size of 2 KB uses at most 2 MB per second of data transfer bandwidth. Because each I/O request occupies a disk completely for the duration of its execution, the way to maximize I/O throughput for I/O-request-intensive applications is to maximize the number of disks that can be executing requests concurrently. Clearly, the largest number of concurrent I/O requests that can be executed on a volume is the number of disks that contribute to the volume's storage. Each application I/O request that "splits" across two stripe units occupies two disks for the duration of its execution, reducing the number of requests that can be executed concurrently and thus the efficiency of I/O response.

Therefore, try to minimize the probability that I/O requests "split" across stripe units in I/O-request-intensive applications.

The following factors influence whether an I/O request with a random starting address will split across two stripe units:

  • The request starting address relative to the starting address of the storage allocation unit (the file extent)

  • The size of the request relative to the stripe unit size

Most database management systems will allocate pages in alignment with the blocks in a file, so that requests for the first page will almost never split across stripe units. However, database requests for two or more consecutive pages may split across stripe units. In this case, larger stripe unit sizes reduce the probability of split I/O requests. However, the primary objective of striping data across a volume is to cause I/O requests to be spread across the volume's disks. Too large a stripe unit size is likely to reduce this spreading effect.