There was never any doubt it would happen, but the argument that Hyper-V is “just a hypervisor” is fast becoming spurious. Last week, Microsoft added fault tolerance to its virtualization repertoire with the announcement that it had entered into a development and marketing agreement with Marathon Technologies.
In many ways, the news isn’t all that surprising. It was, after all, only a matter of time before Microsoft began partnering to grow an infrastructure around Hyper-V.
Marathon’s everRun software family has been in the fault tolerance and high availability market since 1993. As virtualization gained traction, it began working its way into virtual environments, and last March it went whole hog and released a purely virtual play, everRun VM, a version with XenServer underpinnings specifically designed for virtual environments.
With this move, everRun goes beyond Xen and deeply into to Hyper-V territory. Marathon’s deep coupling with Citrix and thus its design capability make it an ideal partner for Hyper-V.
Currently, everRun supports Windows server 2003. Under the terms of the agreement, support for Windows Server 2008 will be added in the second quarter, Marathon President and CEO Gary Phillips told ServerWatch. But the real value for organizations is that everRun will be put on top of Hyper-V, “enabling customers to cross-leverage” the capabilities of both applications in a seamless environment.
Today, Windows Server 2008 customers can use failover clustering. This comes standard with the Enterprise and Datacenter Editions of Windows Server 2008 and is designed to eliminate single points of failure. Component- and system-level fault tolerance will be available via everRun Component Level Fault Tolerance in the second quarter of 2009. This level of fault-tolerance is for applications with a requirement for little or no downtime or data loss. In all cases, Marathon will provide the first line of support for everRun.
EverRun Chief Technology Officer Jerry Melnick explained the terms of the agreement concisely in a guest post on Microsoft’s Windows Server Division WebLog.
In the post, Melnick talks about how everRun mitigates Window’s inherent unsuitability for fault-tolerant computing, despite its increased application in this manner. This is a driver because of the following:
First, more customers are relying on Windows Server to run their mission-critical applications, and the number of these applications is increasing. Second, with the growing popularity of server virtualization (where applications are being consolidated onto fewer servers) the impact of downtime is often magnified.
This brings us to the second part of the agreement. Engineering teams from the companies will work together to build a platform in Windows for ISVs and Marathon to deliver fault-tolerant solutions for other ISVs to work on, Mike Schutz, director of product management, Windows Server Division at Microsoft told ServerWatch. Ideally, this will streamline setup and installation and seamless integration between levels of availability.
It’s not hard to see the synergy in this alliance. It’s also not a stretch to wonder if Marathon is an acquisition target. It is, after all, a venture-backed company, and going public these days is a far less profitable path than being purchased.
From a technological perspective, it’s a decent fit. Marathon has eschewed VMware’s platform in favor of those from Microsoft and Citrix, and is successfully building a customer base without supporting the environment. Should this become a trend with other ISVs, it will not bode well for VMware.
Microsoft remains behind the eight-ball when it comes to a virtual infrastructure, and ramping up via partnerships and acquisitions may be far less costly and more expedient than building one from scratch. It’s more than likely that similar deals will come out of Redmond as Microsoft expands its footprint.
Amy Newman is the managing editor of ServerWatch. She has been covering virtualization since 2001.
This article was first published on ServerWatch.com.