HP pits Matrix against Cisco's California
'We need blades. Lots of blades'
Mark Hurd, in a black leather overcoat, wearing dark shades, standing in a white room that stretches out to infinity. "We need blades. Lots of blades," he mutters.
...And suddenly there are racks and racks of servers, bigger than a Google data centre, hotter than a nuclear simulation, humming away, ready to take on Cisco Systems and its virtualised "California" blade system in a battle for the data center. In this case, though, the Matrix is the weapon, and one with a virtualisation bent, not the virtual world that needs to be destroyed.
Today, HP is announcing something called the BladeSystem Matrix and its Matrix Orchestration Environment (MOE), which is an amalgam of tools that HP has acquired and created from scratch to provision and manage server environments in a more autonomic fashion than has previously been possible.
Cisco is doing a blast-from-the-past by referring to its integrated blade server and networking gadget as a system. The California system is not a diet or a way of life, but a blade server with integrated switching and management located in the switches, formally known as the Unified Computing System.
HP could not just roll out a new blade management tool that is a mix of its Insight Management tools (which it gained through its $25bn Compaq acquisition in September 2001) and its server provisioning and patching tools (from its $1.6bn Opsware acquisition in July 2007) and leave it at that. HP wants drama, like the kind that Cisco is generating. But, unless HP decides to buy Sun Microsystems or something crazy like that, Cisco and Sun are going to be getting most of the drama points in the systems racket for a while. Not the BladeSystem Matrix.
That doesn't mean that an integrated system is a bad idea. With the economy being in the dumps, any technology that allows data centres to be run with fewer people while handling more workloads is going to be something that is popular with the CEOs and CFOs. Cisco is coming at the problem from the networking side and extending into servers, while HP has been coming at it from the server side and extending out to networking for the past decade.
The Matrix system software, according to Gary Thome, chief strategist for infrastructure software and blades at HP's Enterprise Storage and Server unit, is the culmination of a decade of work, which started when HP began talking about utility computing before the Compaq deal. The software weaves together many different technologies, but the new bit - and perhaps the best bit - is a new graphical templating environment. This allows administrators to build a pictorial representation of a software stack as it is running on a multiple-tier collection of servers, then provision this n-tier stack onto real blade servers. It can then be deployed in a matter of days, instead of a month or more across multiple teams (servers, storage, networking, software, and so forth) who all have to sign off on it.
Such graphical tools have been around forever, of course, but they are created by managers and programmers who then hand the pictures to a human being, who then configures and tests the software.
"Matrix is the world's first push-button data center," says Thome. "Or you can call it a cloud in a box. The point is, it converges all of the infrastructure and management together."
That is, of course, the message that Cisco is peddling with the California blade setup. It converges servers and networking into a single, compact box that has virtual and real switches and can be managed with the same tools that Cisco network engineers are already used to. Cisco has not been very clear about storage, beyond saying that EMC and NetApp arrays will plug into the converged switches, which support network and storage over the same switches.
Matrix vs. California...or just go with Liquid
Unified computing - the complete solution (which seems to elude the two big vendors playing in the space today) is a very disruptive force that will fundamentally change how data centers are designed/operated, as well as who the technology leaders will be moving forward. But, the true destructive benefits of the approach are being held back by the big players. Why?
Well, my opinion is that if you're a vendor worried about what to do with the boat loads of '90's era infrastructure that you've manufactured and sold to customers as recently as last month, you've got a problem. Do you want to explain why you sold something that was something less than the optimal solution available to your customer? Of course not. So, your best course of action is to slow the roll-out of unified computing in order to wring the last bit of revenue out of the old stuff that you can while constructing an on-ramp for your customers with the speed of a road crew.
Data center architects who are interested in building and investing into forward looking infrastructure have learned time and again that during water shed moments such as these, the best solutions come from independent vendors who are unencumbered with the past. Cisco used to be one of those companies...20 years ago. Today, look to independent company's such as Liquid Computing (www.liquidcomputing.com), the only vendor that's actually shipping product and has real customers in production.
A fully featured system can see, manage and control compute+network+storage. Anything less is just an on-ramp.