Problem solve Get help with specific problems with your technologies, process and projects.

Channel Explained: Data center design for networking VARs

Networking VARs are under pressure to get into selling the data center -- and that means understanding and selling data center design and services.

Although most networking VARs have traditionally structured their business models around selling networking components and software licenses, that approach is starting to change. Major industry players such as Cisco, ProCurve and Brocade are now allowing VARs to transition from just selling networking components to selling data center components and design. This is both good news and bad news for VARs.

The good news

The good news is that offering full data center design services to your customers opens up new revenue streams. Sure, most of your customers probably aren't going to need a whole new data center any time soon, but you can help your customers figure out what new components they are going to need to meet IT requirements, and you can help them to integrate those new components into their existing data center in an optimal manner. You may also be able to position yourself to help customers choose data center layouts and components that will enable them to scale their existing infrastructure as their needs change over time.

The bad news

The bad news is that even if you don't want to get into the business of designing data centers, you may have to. You will probably find that your competitors are beginning to offer data center design services along with hardware and computing. Since customers tend to like one-stop shopping, you will have to offer the same services as your competitors do in order to retain business.

Furthermore, now that many of the major networking vendors have expanded to offer data center portfolios that include both computing and hardware, they will ask their VARs to sell these technologies along with data center design services. Cisco, for example, is infamous for asking its larger partners to sell end-to-end solutions. Now that the company is asking partners to sell its new server and Unified Computing System, those VARs will have to flesh out their design and services offering as well. But networking partners have a lot to learn -- and sometimes that means starting with the basics.

More on data center design
Tech Watch Analysis: Data center containers promise easy deployment, management
Power and cooling considerations for data center network design   

Building the network infrastructure in your data center  

How to monitor and manage your data center network  

Data center power and cooling study guide

Network cabling

To give you an idea of what I'm talking about, let's discuss something really simple: cabling. I'm sure that -- as a VAR -- you have a good working knowledge of the types of cable that are available and what the bandwidth and distance limitations are for each cable type. For example, Cat 5, 5E, 6, and 6A can all span distances of up to 100 meters (328 feet), but standard Cat 5 cable is rated only for speeds of up to 100 Mbps. If you want to use gigabit speeds, you will have to use Cat 5E or a higher-rated cable. Cat 6 is rated for up to 3 Gbps, while Cat 6A is rated for 10 gigabit speeds.

And there is more to cabling than knowing how far a cable span can run or what type of cable will support specific network speeds. For instance, one potential issue is the local fire codes. Most localities require the use of plenum-grade cables in walls, ceilings and crawlspaces. Plenum cables are designed to be fireproof and not to give off toxic fumes if they get hot. Although the requirement to use plenum cable in walls, ceilings and crawlspaces is pretty common, fire codes vary from one locality to another, and it is your responsibility to make sure that your installation meets code.

Another issue that you will have to consider is the weight of the cable. I once knew someone who damaged a drop ceiling by running an excessive amount of cable without supporting it.

Power management

If you are going to begin offering data center design services, you will also need to have an understanding of power management. For instance, you must be able to determine how much power each device is consuming and how much of a load you can safely put on each electrical circuit.

It is also important to realize that sometimes power management is a bit more involved than just plugging a server into a UPS. For instance, some of your customers may want to use dual power as a way of eliminating a single point of failure. That way, if one source of power fails, there is always backup to avoid downtime.

Another power management technology that it is useful to know about is Power over Ethernet (PoE). POE is useful in situations in which a component is not close to an electrical outlet. For example, I once had a client who had a wireless access point that was mounted to the ceiling of a large warehouse. There were no electrical outlets on the ceiling, so POE proved to be the best way to power the device. POE can also be used to reduce cable clutter. Most vendors now have data center components that are PoE ready.

Rack layout

Another aspect of data center design that a VAR may want to consider is rack placement. Rack placement in a data center is something of an art form. First and foremost, racks need to be arranged in a way that allows all of the networking components to physically fit into the data center. Racks are also typically arranged so that components that serve a related purpose are located in close proximity to one another. Keep in mind, though, that the appropriate rack layout is going to vary, depending on the components that you are mounting. Not every component manages airflow in the same way.

Ultimately, racks must be positioned in a way that promotes good airflow. For example, if an organization has a number of rack-mount servers that have an air intake in the front and an exhaust port for hot air in the back, you would not want to arrange parallel racks of these servers in a way that had all the servers facing the same direction. In that case, the hot air that is expelled from the first row of servers would be sucked into the second row of servers, where the air would be further heated before making its way into the next row of servers.

To prevent this from happening, data centers often use a layout called hot-row/cool-row. The idea behind this design is that if the servers all have exhaust ports on the back of the chassis, the servers should be placed back to back. That way, the front side of each server (where the cool air enters the server) is not facing another server's exhaust port.

Although making good use of floor space and using a design that allows servers to be properly cooled is important, you must also arrange server racks so that both the front and back sides of the components in each rack are accessible, as are the cables that connect to those components. You never know when you are going to end up having to run additional cabling to a rack, so you have to make sure that the data center's physical layout makes it easy to do so.


Ultimately, adding data center design services is probably going to be profitable for most VARs because it opens new revenue streams. At the same time, though, additional training will probably be necessary. As networking vendors ramp up their data center offerings, this is the time for VARs to begin prepping for action.

Brien M. Posey, MCSE, is a Microsoft Most Valuable Professional for his work with Windows 2000 Server and IIS. Brien has served as CIO for a nationwide chain of hospitals and was once in charge of IT security for Fort Knox. As a freelance technical writer, he has written for Microsoft, CNET, ZDNet, TechTarget, MSD2D, Relevant Technologies and other technology companies. You can visit Brien's personal website at

Dig Deeper on Managed network services technology

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.