Types of storage devices

Edit online

You can use various storage devices with IBM Storage Protect to meet specific data protection goals.

Storage devices and storage objects

The IBM Storage Protect server can connect to a combination of manual and automated storage devices. You can connect the following types of storage devices to IBM Storage Protect:
  • Disk devices that are directly attached, SAN-attached, or network attached
  • Physical tape devices that are either manually operated or automated
  • Virtual tape devices
  • Cloud object storage
IBM Storage Protect represents physical storage devices and media with storage objects that you define in the server database. Storage objects classify available storage resources and manage migration from one storage pool to another. Table 1 describes the storage objects in the server storage environment.
Table 1. Storage objects and representations
Storage object What the object represents
Volume A discrete unit of storage on disk, tape, or other storage media. Each volume is associated with a single storage pool.
Storage pool A set of storage volumes or containers that is the destination that is used to store client data. IBM Storage Protect uses the following types of storage pool:
  • Directory-container storage pools
  • Cloud-container storage pools
  • Sequential-access storage pools that are associated with a device class
  • Random-access storage pools that are associated with a device class
Container A data storage location, for example, a file, directory, or device.
Container storage pool A storage pool that a server uses to store data. Data is stored in containers in file system directories or in cloud storage. Data is deduplicated, if necessary, as the server writes data to the storage pool.
Device class The type of storage device that can use the volumes that are defined in a sequential-access or random-access storage pool. Each device class of removable media type is associated with a single library.
Library A storage device. For example, a library can represent a stand-alone drive, a set of stand-alone drives, a multiple-drive automated device, or a set of drives that is controlled by a media manager.
Drive An object of a tape library device that provides the capability to read and write data to tape library media. Each drive is associated with a single library.
Path The specification of the data source and the device destination. Before a storage device can be used, a path must be defined between the device and the source server that is moving data.
Data mover A SAN-attached device that is used to transfer client data. A data mover is used only in a data transfer where the server is not present, such as in a Network Data Management Protocol (NDMP) environment. Data movers transfer data between storage devices without using significant server, client, or network resources.
Server A server that is managed by another IBM Storage Protect server.

The administrator defines the storage objects in the logical layer of the server, as illustrated in Figure 1.

Figure 1. Storage objects
Illustration shows types of physical layer storage devices and logical representations

Disk devices

You can store client data on disk devices with the following types of volumes:
  • Directories in directory-container storage pools
  • Random-access volumes of device type DISK
  • Sequential-access volumes of device type FILE
IBM Storage Protect offers the following features when you use directory-container storage pools for data storage:
  • You can apply data deduplication and disk caching techniques to maximize data storage usage.
  • You can retrieve data from disk much faster than you can retrieve data from tape storage.

Physical tape devices

In a physical tape library, the storage capacity is defined in terms of the total number of volumes in the library. Physical tape devices can be used for the following activities:
  • Storing client data that is backed up, archived, or migrated from client nodes
  • Storing database backups
  • Exporting data to another server or offsite storage
Moving data to tape provides the following benefits:
  • You can keep data for clients on a disk device at the same time that the data is moved to tape.
  • You can improve tape drive performance by streaming the data migration from disk to tape.
  • You can spread out the times when the drives are in use to improve the efficiency of the tape drives.
  • You can move data on tape to off-site vaults.
  • You can limit power consumption because tape devices do not consume power after data is written to tape.
  • You can apply encryption that is provided by the tape drive hardware to protect the data on tape.

Compared to equivalent disk and virtual tape storage, the unit cost to store data tends to be much less for physical tape devices.

Virtual tape libraries

A virtual tape library (VTL) does not use physical tape media. When you use VTL storage, you emulate the access mechanisms of tape hardware. In a VTL, you can define volumes and drives to provide greater flexibility for the storage environment. The storage capacity of a VTL is defined in terms of total available disk space. You can increase or decrease the number and size of volumes on disk.

Defining a VTL to the IBM Storage Protect server can improve performance because the server handles mount point processing for VTLs differently than for real tape libraries. Although the logical limitations of tape devices are still present, the physical limitations for tape hardware are not applicable to a VTL thus affording better scalability. You can use the IBM Storage Protect VTL when the following conditions are met:
  • Only one type and generation of drive and media is emulated in the VTL.
  • Every server and storage agent with access to the VTL has paths that are defined for all drives in the library.