HP StorageWorks 70 HP StorageWorks Replication Solutions Manager 4.0.1 user gu - Page 438
Redundancy level Vraid, Snapclones
View all HP StorageWorks 70 manuals
Add to My Manuals
Save this manual to your list of manuals |
Page 438 highlights
• Cannot be a snapshot or a mirrorclone. See virtual disks types. • Cannot have a mirrorclone. This restriction is controller software version dependent. See controller software features - remote replication. • Must be in a normal operational state. See resources operational states. • Must use mirrored cache. See virtual disks cache policies. • Must have the same presentation status as other virtual disks in the DR group. With some versions of controller software, the virtual disk must be presented to a host. See storage systems controller software features - remote replication. See also virtual disks presentation. Redundancy level (Vraid) Redundancy level (Vraid) refers to the method and extent of RAID data protection for a virtual disk. Vraid levels are defined as follows: Type Redundancy Technique Remarks Vraid1 High Mirroring Optimized for redundancy and speed, but requires twice the disk space of other levels. Vraid5 Medium Striping and parity Provides a balance of redundancy, speed, and disk space. Vraid0 None Striping Optimized for speed and disk space, but provides no redundancy. IMPORTANT: HP does not recommend using Vraid0 when high availability is required. Vraid is the HP StorageWorks implementation of RAID on Enterprise Virtual Arrays. However, unlike traditional RAID, all Vraid levels distribute data blocks across multiple physical disks in the disk group. Once a virtual disk or container is created, its Vraid level cannot be changed. Snapclones Snapclone replication instantly creates an independent, point-in-time copy of a virtual disk. The copy begins as a fully allocated snapshot, and then automatically becomes an independent virtual disk. The copy is called a snapclone. See also virtual disks snapclone FAQ and snapclone guidelines. The snapclone property indicates whether a virtual disk can be locally replicated using the snapclone method. Values are: • Yes. The virtual disk complies with snapclone guidelines. Snapclone replication can be performed. • No. The virtual disk does not comply with snapclone guidelines. Snapclone replication cannot be performed. Preallocated snapclones Preallocated snapclone refers to a snapclone that is created by copying data from a source virtual disk to a container and immediately converting the container into a virtual disk. Compared to a standard snapclone, creating a preallocated snapclone is faster. In cases where host I/O must be suspended, the improved speed of preallocated snapclones reduces the time that a host application is suspended. When a preallocated snapclone is created, the source virtual disk write cache must be flushed before replication is started. (See cache policies write cache.) This ensures that the source virtual disk and snapclone copy contain identical data. The following table shows how a write cache flush is implemented. IMPORTANT: When using jobs or the CLUI, you must explicitly ensure that write caches are flushed. 438 Virtual disks