(This article originally appeared in the October 2009 issue of Unified Communications.)
At first glance, one might think that Panduit is solely a huge manufacturer of the kind of hardware found in data centers and the network itself: cables, connectors, cabinets, racks, etc. In actuality, Panduit is also a developer and provider of industry-leading solutions that help customers optimize their physical infrastructure. These tailored, end-to-end, unified physical infrastructure-based solutions give enterprises the ability to connect, manage and automate communications, computing, power, control and security systems for a smarter, unified business foundation.
Panduit is able to formulate these sophisticated, unified solutions across a wide range of deployments thanks to their tremendous expertise in global manufacturing, logistics, and e-commerce capabilities, along with their global ecosystem of consultants, integrators, contractors, a network of distribution partners, and close technical relationships with major industry systems vendors.
Recently, I delved into the art of UPI-based solutions by talking with Panduit’s Vice President of Global Marketing, Vineeth Ram.
RG: Tell me how Panduit is evolving into being able to provide more of an integrated solution to customers involving the complete physical infrastructure.
VR: We think of ourselves as a physical infrastructure category leader. We offer a complete solution. We connect, manage and automate five types of systems: power, control, compute, communications and security. These can be anywhere – in a building, in a data center, in your enterprise, on a factory floor. In today’s world, the big story is that there’s more and more need for these different systems to come together as we all must communicate and collaborate.
And, as IP-based communications and various technologies change and evolve, you have more of a pronounced need for communication and integration at the system level, which introduces many requirements into the very physical infrastructure that is the foundation where you connect, manage and automate these systems. That’s why we believe it’s becoming a lot more relevant to be a physical infrastructure leader.
Key challenges still exist, for instance in the data center, concerning power, space, cooling and performance, in addition to maintaining mission-critical availability, reliability and agility. These are all tied to the need to provide real-time information even in the case of mobile access, with IP communications functioning across multiple kinds of environments.
RG: So you’re really pulling these things together into full solutions, rather than just putting together a cabinet and some components.
VR: Well, the five types of systems I mentioned – power, control, compute, communications and security – are starting to converge. As this occurs, there’s a whole range of problems that must be addressed together.
For example, as each new computer becomes faster than previous models, they generate more and more heat, so power and cooling is a big issue today, and it also impacts availability.
At the same time vendors are enabling consolidation projects. They say, ‘Don’t use 50 boxes when you can do the same thing with 10 of these new boxes.’ However, the heat generated for each of these new devices now goes up significantly. Installing more capacity into the same data center space also means that you need greater amounts of high-speed bandwidth for high-speed networking required by consolidation, virtualization and cloud computing.
This all makes for a great story, since you can say that you’re saving costs and improving utilization at the systems infrastructure level, but it introduces a whole range of requirements into the physical foundation below that. There’s a challenge managing all of that, so that all of the passive cooling capability becomes very critical.
Essentially, what’s really happening is that, when you talk about these challenges, such as power, cooling, space and high speed networking performance, you’re going beyond the siloed domains of solving a problem. For example, take somebody who makes a copper or fiber cable and associated connectors. They cannot directly solve your power and cooling problems; they are simply an element of the solution. Somebody else makes your cabinet, somebody else makes your patch panels and intelligent management of your infrastructure. Somebody else is making all of your electrical pathways and doing the grounding and bonding between devices and an earth ground. The industrial tradition has thus been a ‘piece-part’ approach, which unfortunately doesn’t address your power, cooling or space kinds of challenges.
On a solution level, let’s say you’re a Cisco (News - Alert) customer deploying a large top-of-rack system architecture. You have a certain logical topology and you have your Cisco Nexus switches and MDS SAN network, and so forth. Now, how do you go and deploy that in an optimized manner in a data center, so that the physical deployment optimizes and solves the challenges of power, cooling, speed, space and performance? And how do you future-proof your investment?
Panduit’s approach is to say, ‘We can put together a complete reference architecture around all of these different pieces, but brought together in an integrated manner that solves these challenges and optimizes the system.’ We do that by collaborating closely with vendors such as Cisco, IBM, EMC (News - Alert), VMware, and a whole range of technological leaders, and put a complete deployment together for the physical foundation. We integrate everything from the cables, the connectivity, the racks, the cabinets, the pathways, the grounding and bonding and the intelligent physical layer and software management. That all becomes very critical when you want to do virtualization because you’ve got to map out the actual application – which application is running on which virtual machine, sitting on which physical server, and then you need to know on which cabinet and rack it’s positioned.
After all, you need to be able to provision the resources rapidly, and you want to be assured that there’s going to be power at the specified locations, and at the same time you don’t want somebody to go pull out a plug with a mission-critical workflow moving through that location.
RG: So you’re taking best-of-breed components and assembling them into an intelligent solution to achieve, among other things, a smaller, greener footprint.
VR: Yes, a more scalable core architecture. Previously, everything until recently was a silo. The facilities people don’t talk to the IT department, and IT doesn’t talk to the network guys, and on and on. Everything was a separate silo.
Now we talk about the unified physical infrastructure. Systems are converging. When we develop best practices and reference architectures, and a methodology for a full solution, now it becomes much easier for a facilities person to say, ‘Oh, you’re talking about this type of Cisco Nexus deployment. I know exactly what the topology looks like to deploy that kind of physical solution that’s scalable.’ He doesn’t need to know everything about networking, but he’s got the answer to deploy that from a physical standpoint. That’s the benefit that Panduit brings into play in this market. We’re trying to help break down the silos with a unified, integrated, physical infrastructure that solves these challenges. It’s more of a horizontal approach, but in some sense it’s also vertical because what we do is to spend a lot of time understanding technology.
For example, when Cisco recently launched their 7018 Nexus switch, Panduit designed a specially-defined cabinet with all the necessary cable management pathways, it really helped Cisco solve some of the power and cooling challenges, since it was specially designed for that. We optimized the cabinet design, starting with a 32-inch cabinet and settling on a 40-inch design that worked beautifully. So we know the best practices around that, since we designed and built it. It’s fully integrated with cable management. Now, we can go and translate many of the logical parameters into what that means from a physical deployment perspective.
We work similarly with companies such as IBM (News - Alert) with their blade server technology, EMC and their storage networking technology, and Rockwell Automation from a factory automation perspective. On the building automation side, we work with Johnson Controls and other leaders. This approach gives us a complete physical infrastructure foundation to address the challenges being faced every day.
In the past you could say, ‘Don’t worry about the cabling infrastructure, just go deploy the switch and somebody else will cable it for us.’ Those kinds of discussions were fine when your bandwidth was 1 Gbps or less. The network was almost an afterthought in terms of physical layer 1. Today, there are very powerful, tight and intricate network connections, because there’s so much efficiency being driven in the data center environment. Cisco talks of the top-of-rack architecture, and we must now use new types of interconnects, to support the latest high data rate, latency, power and distance requirements. That’s why we now have things such as Small Form-Factor Pluggable SFP+ connectors, with logic literally built inside to allow equipment to determine what type of cable it is.
Whatever physical structure you deploy, it has a useful life of perhaps 10 to 15 years, and that spans about three to four technology refresh cycles involving systems technologies. That is why we at Panduit make deployments as scalable and future-proof as possible.
We’re always one step ahead of the curve, ready for what’s coming down the road.
Richard Grigonis is Executive Editor of TMC (News - Alert)�s IP Communications Group. To read more of Richard’s articles, please visit his columnist page.
Edited by Erik Linask