HDS adds failover clustering
Baby step towards 'future of storage'
Replicating features that NetApp and IBM already support, Hitachi Data Systems has added failover clustering to its enterprise-class USP-V storage arrays.
The USP-V (Universal Storage Platform - Virtualising) is a virtualising storage controller that can combine both Hitachi and third-party storage arrays into a single logical pool of capacity. As expected, HDS has announced its High Availability Manager software for the USP-V, which will automatically failover operations to the second member of a clustered pair of USP-Vs
In the event of a controller failure the second controller instantaneously takes over operation and continues data access with no application downtime. HDS says that the capability is worthwhile for data migrations from an older array to a new array as this can be achieved with no downtime.
Cache coherency in the cluster is arranged through local mirroring for the controllers and an external quorum disk.
Failover to remote sites is supported through the use of Hitachi Universal Replicator and Hitachi TrueCopy software. HDS also now supports IBM FlashCopy technology. When used with Hitachi Universal Replicator software, HDS says it improves business continuity and disaster recovery strategies in both 2-data centre point-to-point operation and 3-data centre multi-target configurations.
HDS says its new software provides the "industry’s first continuous availability solution for both internal storage and externally attached heterogeneous storage."
HDS blogging representative Tony Asaro writes: " What other Enterprise-class storage system offers a solution that ensures your applications never have to be brought offline? This is extremely compelling, unique and valuable to customers. Hitachi High Availability Manager redefines high availability in the data centre and no other storage system can offer this capability."
IBM's SAN Volume Controller (SVC) virtualises storage from IBM and other vendors behind pairs of SVC Nodes (I/O Groups) with failover between nodes in a pair and support for non-disruptive migration.
NetApp's V-Series virtualises third-party storage and can be configured in a clustered pair with what NetApp says is seamless failover, with no detectable interruption and greater than 99.99 percent availability.
It is expected that HDS partners HP and Sun, which supply USP-V-based products, will sell the High-Availability Manager software.
The product will be available in the fourth quarter of the year. Pricing will relate to that of the host USP-V controller on which it is deployed.
HDS billed this announcement as the future of storage. Its HAM is an evolutionary step in the development of the USP-V, but not a development of its architecture that is scalable either in performance or capacity up to higher levels. Only a little step forward then. ®
Sponsored: RAID: End of an era?