I had a great conversation with Jerry, the CTO of Marathon Technologies, a couple of days back. What I really find very interesting is that they are committing to Open Source and Xen technology. They are teaming up with Citrix to bring virtualization and the HA (High Availability) to the masses.
Jerry's profile (from the site):
Jerald H. Melnick, Chief Technology Officer
Jerry is responsible for Marathon’s technology roadmap that is driving the convergence of high availability and virtualization technologies. Under Jerry’s leadership, Marathon is developing the first-ever fault tolerant-class software solution for virtual environments. Built in partnership with XenSource, this cutting edge product was awarded “Best New Technology” at VMworld 2007. Before joining Marathon, he held executive positions at PPGx, Inc. and Belmont Research as well as management and technical roles at Digital Equipment Corporation, where he was responsible for the development and deployment of mission-critical platforms to support enterprise-computing environments. He led a variety of system and product development efforts in the area of operating systems, network communications, database systems, and computing languages. Jerry holds a BS in chemistry from Beloit College and graduate work in computer engineering and computer science at Boston University.
Here is the Q&A:
- Please tell us about Marathon?
Marathon is the world’s only provider of fault-tolerant class high availability software (and services) for server virtualization, specifically designed to protect Windows applications running within a Citrix XenServer Enterprise infrastructure. We’ve been building and deploying availability solutions for more than 14 years and thousands of customers now use our products to protect their most critical business applications. Our technology is novel, employing more than a dozen patents. We’ve developed a next-generation architecture that achieves industry leading availability in a format that is extremely easy to configure and administer. Server virtualization is a cornerstone of the technology.
- Please tell us a bit about your product line?
The product line is named everRun. everRun delivers a complete range of automated, fault tolerant-class availability and disaster recovery software solutions. There are several flavors depending on the target environment but generally speaking, everRun is architected to protect one or more virtual machines from unplanned outages.
- How does Marathon Technologies complement a Virtual Infrastructure?
everRun is tailor made for the virtual infrastructure. First off, server consolidation creates the well-known “all your eggs in one basket” syndrome - a single point of failure that has the potential to take down all of the applications hosted by a server. everRun clearly addresses this important pain point because it protects the virtual machines from server component or host failures.
At another level, we see a gap between the promises of server virtualization and the need for what I would call “real” availability solutions. The lack of mature availability solutions for virtualized environments is impeding the use of virtualization for the most important applications. everRun hardens the virtualization environment so a company can leverage the benefits of virtualization across the broadest spectrum of applications including their critical applications.
We also talk to a lot of customers that have applications they “should” protect. But, on a physical platform it may be too difficult or expensive to do so. As they move these applications to virtual environments, everRun allows them to very simply turn on protection in a very seamless and cost effective manner. So, in all of these ways, we’re actually broadening the use cases for virtualization in a very complementary fashion.
- How can I be assured that my Virtual Machines are always live and kicking?
This is precisely what everRun does. It assures that your protected VMs are always alive. By creating and managing redundant environments across a XenServer pool, everRun is continually “self-checking” the environment to make certain that the necessary resources are available and operational. If any server, component or IO pathway fails, everRun takes immediate and automatic actions to transparently reconfigure operations to maintain application execution and client connectivity. So, your application is always available.
Just as importantly, if a failure in a redundant component is identified, everRun notifies an administrator so that a repair can be made. This provides the assurance that recovery from failures is always possible.
- What are your efforts around Open Source, is it part of your strategy, given that Open Source is getting a lot of attention from the business community?
Open source Xen allows us to innovate and rapidly build high quality products that meet our customer’s needs. So, certainly, OSS is an important part of our strategy. So, we use Xen as a platform and environment to deploy our technology and where we need enhancements from the platform we have the capability to enhance it. That is certainly important for the type of advanced technology we provide. And of course, for support and all of the usual reasons, it is critical that we base our shipping product on a well tested and mature distribution of Xen. That’s why we’re working closely with Citrix today to put this all together.
- Who are your direct competitors?
It may sound a bit self-centered but we really don’t have a direct competitor. There are several vendors out there who have failover products that help recover data (so called replication & failover products) but none (that we’re aware of) that are automated, prevention-oriented, and able to provide a desired level (selectable) of availability across a virtual infrastructure whether protecting a single virtual machine or many.
- Why would you want me to deploy everRun instead of, say Neverfail? What makes you stand apart?
The comparison is somewhat of an apples and oranges proposition.
At the root, Marathon is in the failure “prevention” business versus “recovery” – we do that too but philosophically we strive to prevent failure at the onset and if a failure does occur, we enable the platform to automatically remediate the issue while riding through an event without disruption to the end user.
everRun is non-intrusive and layers into a Citrix XenServer Enterprise environment (and an otherwise standalone Windows Server 2003 environment too) without having to touch a VM (e.g., the guest OS or the application).
everRun also has a unique geographic separation capability that allows the abstracted hardware to be in different geographic locations (as opposed to in the same rack in the same data center) or in a pool of servers in a single/primary data center.
- Can you tell me a bit about lockstepping?
The simplest description of lockstepping is when two servers (CPU, Memory, I/O) are joined and initialized to the same state during system start-up. Once joined, the two servers execute the same instructions and act identically. From the outside, the two systems look and behave like a single server. But the redundant hardware and execution environment provide all the needed resources to transparently handle a server or component failure.
Lockstep logic continually monitors data and events across both machines to ensure that lockstep operations are maintained. In the event of a fault on either system, the fault will cause a state difference between the two servers which the systems monitors detect and, in everRun’s case, automatically remediate the condition. The primary benefit of lockstep is that if there is a server fault, the surviving server can ride through (no application downtime) the event without user interruption (or data loss) to the processes taking place on the server eliminating the need, complexity, and time necessary to manage a failover event.
- What is v-Available?
v-Available™ is a Marathon Technologies initiative designed to educate the industry about the need for fault tolerant-class availability for virtual environments. Marathon launched the v-Available initiative to fill a big gap left by today’s high availability offerings for virtual environments. Virtualization brings many benefits – but it also dramatically increases the need for rock-solid availability. Why? Because server consolidation can result in the server becoming the single point of failure for multiple applications, so the implications of downtime are much greater.
- What are you doing to get more visibility, given that Virtualization is hot, and everyone is going for that big $15-20 Bn market?
Lately, Marathon has been doing rather well in this regard as evidenced by the validation we’ve received from CIO Magazine and Best of Show at VMWorld. Our best visibility has been, and will likely continue to be, word-of-mouth from happy customers. It also helps that we’re laser focused on being the “availability” ingredient that makes Citrix XenServer Enterprise not only a good choice, but a great choice for those companies looking to virtualize their environments while virtually eliminating the needless risk of unplanned downtime.
- What is the biggest deployment you ever did (you don't have to name your client) and what were the challenges?
The direct answer to your question is that our largest customer has deployed over 40 everRun licenses in their environment for a variety of applications, ranging from Exchange to SQL to custom-developed (broadcast media) applications.
We also have large deployments in manufacturing, pharmaceuticals, finance, and local government. In the majority of these deployments, customers find it extremely easy to setup and configure our system. Typically, it’s a 2-3 hour installation including the Windows environment. Once it’s set up, everything else is automatic. There’s really no policy to define. You install the application just as you would on any server. Storage is automatically mirrored and networks are redundantly monitored. When a failure occurs, everRun automatically reconfigures and keeps the application running. So, it’s really very easy.
There are some more advanced configurations out there that leverage our disaster tolerant capabilities. This feature (we call it Split-site) allows redundant servers to be placed and protected in separate physical facilities. So, if you lose a facility, we keep your application operating without any intervention. In a few of these deployments, we have helped our customers configure the environment and network for deployment. Once this has been done, our customers haven’t really needed us much to expand or reconfigure.
But, as you can see, for the most part, this is a hands-off technology. It makes it easier for all of us and it’s really the very reason it is so reliable.
- What are your future plans and near-future plan in 2008?
Here again, our plans are rooted in our ability to focus e.g., being the “availability” ingredient that makes Citrix’s XenServer Enterprise not only a good choice, but a great choice for those companies looking to take advantage of the virtualized environments while practically eliminating the needless risk of unplanned downtime.
Expect Marathon to continue to deliver products and service that give customers a real choice when setting out to virtualize their environments, without having to compromise on what should be “check mark” availability features and benefits.
- Is there a buyer out there for Marathon Technologies?
For Marathon’s everRun product – well yes, there’s a plethora. For Marathon Technologies, the company, no comment J
Marathonplan to stay long as privately funded firm or will we see an IPO?
Here again no comment other that to say that