Chapter 11: Simplifying Disk Management for Storage Assets

---

A lot is changing in the subject of disk management as a result of the release of Windows Server 2003 (see Chapter 15). Most of the material covered in this chapter is part of Windows Server 2003. What we have to look forward to is greater functionality in the management of storage disks in the field of storage virtualization, as well as better management of data by users courtesy of virtual shadow sets. Further breakthroughs are a result of RAID-type snapshot technology that copies the complete contents of a volume in seconds, enabling transaction-heavy systems such as SQL Server to stay online while the data is being backed up.

Snapshot Technology

Snapshot technology is used now by many vendors. Essentially, a snapshot is an image or copy of a defined collection of data created instantly at a point in time. Copies are made almost immediately within the disk subsystem, despite the size of the volume. The primary use of a snapshot is to facilitate nondisruptive backups. Essentially, the snapshot image becomes the source of backup. After halting the application, the copy takes only a moment to create, so the user should not notice any delay. Implementations of snapshot vary from vendor to vendor. The two primary techniques are copy-on-write (loosely characterized as software based) and split mirror (loosely characterized as hardware based).

Copy-on-Write

In a general sense, copy-on-write can be characterized as software-based snapshots. Primarily, this application is used as server-based software, although the distinction of being software-only may be blurring as copy-on-write is now also being implemented inside intelligent switch or storage virtualization devices. Depending on the vendor, these devices may deploy software or firmware to provide copy-on-write capabilities. Basically, with copy-on-write, the snapshot is a logical copy of the data that gets created by saving the original data to a snapshot index whenever data in the base volume is updated. It provides efficiency by requiring only a fraction of the base volume disk space. Periodic snapshots can be made throughout the day. Sophisticated roll-back capabilities are being developed to instantly roll back a complete volume. Soon users will be able to select the time of the restore point and instantly revert to that moment.

Split Mirror

Split mirror could be referred to as the hardware-based approach to snapshots, but this is not strictly true. Just as RAID can be done via hardware or software, so can split-mirror technology be hardware or software based. Most vendors favor the hardware side, although some have software products that use the split-mirror method. The software involved, though, is primarily proprietary and works only with specific hardware from the same vendor and is included in the price. Split mirror is based on RAID/mirroring technology. In disk mirroring, two or more up-to-date full copies of the data are maintained. Every write request to the original data is automatically duplicated to other mirrors or copies of that data. The mirror may be contained in the same subsystem or may be between different subsystems, although these typically must be of the same subsystem model.

The mirror can be temporarily suspended — also referred to as a broken or split mirror — to create a snapshot or point-in-time copy. The disk subsystem is told to temporarily stop making updates to the mirrored copy, so the data is frozen at the point of the suspension. The split mirror can then be used for the backup process or other purposes.

Mirrors create an instant copy, or snapshot, of the data with the split capability. Unlike copy-on-write, a full data copy is available. In order to keep the disaster recovery copy available, a third mirror is usually established for the purpose of splitting. This requires three entire copies of the data volume to provide the protection and meet continuous processing demands for backup and other development needs. This setup has a primary and secondary realtime copy, as well as a tertiary point-in-time copy of the data. In the more sophisticated systems, the writes have been saved from the point of the suspension, they are applied to the mirror, and normal mirror operations resume; however, in some products, the entire mirror must be rewritten once the mirror is broken. In either case, the original data volume is not affected by breaking the mirror.

Products that utilize the split mirror to provide an instant copy or snapshot include:

• EMC TimeFinder (software included as part of the Symmetrix line of storage hardware)

• Hitachi Data Systems InstantSplit for ShadowImage (Lightning series of storage disk systems)

• HP SureStore Business Copy

• Sun StorEdge Instant Image (9900 server systems)

• XIOtech REDI (MAGNITUDE storage hardware)

• LSI Logic ContinuStor Director

• VERITAS Volume Manager and Volume Replicator

Hitachi Data Systems' ShadowImage technology, for example, has been incorporated into Computer Associates' BrightStor Enterprise Backup (EB) as well as the storage products of several leading vendors to create impressive results on the SQL Server platform.

---