Vision to Support IBM’s HASM Technology in Clustering Software
March 11, 2008 Alex Woodie
Vision Solutions last week announced that it’s supporting the new High Availability Solutions Manager (HASM) that IBM will include with the forthcoming release of i5/OS V6R1. With HASM support in Cluster1, the largest System i shops will have a single solution for managing their entire high availability environments, from logical replication all the way through hardware-based clustering solutions such as HASM.
When it becomes available with i5/OS V6R1 on March 21HASM will provide another way for System i shops to replicate data, objects, and resources as part of a high availability or disaster recovery program.
HASM combines the so-called hardware-based clustering capabilities, Global Mirroring and Metro Mirroring, which IBM has made available for users of its high-end external DS8000 storage area network (SAN) arrays, with the new cross-site mirroring (XSM) “under the covers” object- and data-replication technology that IBM has built into recent releases of i5/OS.
The technology is based loosely on the High Availability Cluster Multi Processing (HACMP) technology that it has supported on RS/6000, pSeries, and System p Unix servers for years, and supports up to 32 nodes. IBM advises that customers with the tightest recovery time objectives (RTOs) and recovery point objectives (RPOs) requirements are the best candidates for full HASM automation; in other words, the biggest financial services firms running 570s and 595s in combination with DS8000 SAN arrays will probably be the first adopters.
By providing a single way to manage XSM and SAN-based geographical mirroring technologies–and a single GUI for performing roll swaps–HASM could be viewed, ostensibly, as the most complete and reliable high availability offering the System i platform has ever seen. At least, that’s how IBM is talking about it.
But, of course, HASM hasn’t yet seen the light of day. It hasn’t yet been put into production, and nobody knows how readily System i shops will adopt yet another new, potentially complex storage technology. If the introduction of past storage and replication technologies–remote journaling, clustering, iASPs, XSM–is any guide, HASM uptake will be slow at first, if only because iASPs have yet to see widespread use.
While HASM competes a bit more directly with IBM’s high availability partners, Big Blue is also counting on the help of its vendor community to educate users on HASM and spread the message of HASM. To that end, Vision Solutions, its biggest and most experienced i5/OS high availability software partner, has stepped up and announced its support for HASM in Cluster1, its flagship cluster management software. It’s the first high availability software vendor to do so.
Alan Arnold, executive vice president and chief technology officer for Vision Solutions, sounded bullish on the combination of HASM with Vision’s high availability clustering offering. “This collaboration and bundled offering sets a new precedence of flexibility and fault tolerance within our industry, protecting the most demanding high availability environments with built-in autonomics and relative simplicity,” Arnold stated in a press announcement.
While HASM doesn’t require a third-party logical replication product like Vision’s Cluster1, the reality is the customers that will adopt HASM first probably already have some form of logical replication in place. After all, these early adopters will be the biggest organizations with the most stringent RPO and RTO requirements. Because Vision already has relationships with a good percentage of these customers going back over a decade in many instances, Vision’s cooperation and participation is welcomed by IBM.
But the advent of HASM also marks a shift in the role of Vision and other high availability software vendors. At this point, Vision’s role appears to be acting as a systems integrator to make sure all the high availability technologies–logical replication, clustering, XSM, Metro Mirroring, and now HASM–play nicely together at customer sites. The company has already embarked on ways to make these hybrid high availability environments easier to implement and manage, and support for HASM is one more link in that bridge.