Data Center Convergence: Fact or Fiction?

Recently, IT departments have had trouble keeping pace with growth in their data centers. In many cases, each time that a new application was built, they had to deploy a new server, storage system, and network switch. With IT becoming a key business enabler and the number of applications swelling, they ended up with a plethora of devices.

November 30, 2010

4 Min Read
NetworkComputing logo in a gray background | NetworkComputing

Recently, IT departments have had trouble keeping pace with growth in their data centers. In many cases, each time that a new application was built, they had to deploy a new server, storage system, and network switch. With IT becoming a key business enabler and the number of applications swelling, a plethora of devices resulted.

To ease the burden, vendors, such as Brocade, Cisco, EMC, HP, and IBM have been building a new generation of data center devices. These products (which go by various names, such as converged systems, converged infrastructure, and unified computing systems) consolidate autonomous products (servers, storage systems, network devices) into single systems. This model promises to reduce costs, speed up deployments, and simplify maintenance.

While there are potential benefits, these products also have some shortcomings. The market for these devices is still developing, and customers may find holes in the infrastructure. "To implement these systems, companies will need to make a significant capital investment, usually one starting in six figures," notes Galen Schreck, principal analyst at Forrester Research. In addition, deployment of these systems often requires that companies revamp their IT operations dramatically. Consequently, a slow rather than rapid ramp-up to these systems is expected.

Since the turn of the millennium, IT departments have been under stress. As companies deployed new applications, the underlying infrastructure expanded - often quite significantly. Even small and midsize businesses found themselves working with tens, hundreds, thousands, or even tens of thousands of network switches, servers, and storage systems.

Improvements in automation, semiconductor technology, and virtualization enabled vendors to build more powerful, cohesive devices. One big potential draw to collapsing server, network, and storage systems into a single chassis is cost reduction in areas like physical infrastructure. Companies no longer have to string cable to connect a server or a storage system to a switch, a change that could reduce cabling by as much as 40 percent. (Estimates are that as much as 15 percent of a data center's IT budget goes into the cabling.) In addition, fewer devices mean less use of power and cooling systems.Another plus is that corporations can speed up IT service provisioning. Because the underlying items are integrated, IT departments can get a new system up and running in days rather than weeks, and adding storage or network bandwidth can be done with a few mouse clicks.

Businesses are also able to streamline management functions. "Since the integrated systems include fewer components, it becomes easier to troubleshoot problem connections," notes Scott Dennehy, a senior analyst at Technology Business Research. 

While there are many potential benefits, few companies are realizing them now. One reason is that the underlying infrastructure is still being put into place. A key component is Fiber Channel over Ethernet, which moves Ethernet into the storage arena. Support for that option is just beginning to work its way out from routers and switches into the server and the storage domains.

Getting all of the different components to work together can be a challenge. "These solutions work best when a company has equipment from a single vendor; it is not easy to integrate systems from different suppliers," states Dennehy.

Management tools represent another nascent area. Vendors need to create new role-based monitoring systems, so server or storage administrators can manage items within their domains and virtualization administrators can see the whole picture.Personnel issues could present a major challenge. The traditional IT infrastructure tends to be siloed: "Usually companies have a server technician, a network administrator, and a storage specialist," says Forrester Research's Schreck. With these tools, the dividing lines among these different groups have become blurry. For example, a VMWare hypervisor could be embedded in a network switch.

So, because of these new systems, IT personnel need to understand and be able to manage a wider variety of devices. Consequently, training will be needed, so employees can fill in any missing areas of expertise. In addition, companies will need to reassess the way their staff is structured and how it operates. Many business practices will have to be redone.

Vendors say that implementing these new systems may enable companies to reduce staff size. Long term, companies could benefit from such changes, but this possibility could hinder productivity in the short term as individuals become concerned about job security.

A select number of companies is starting to put some of the pieces in place, but some time will pass before these systems are widely deployed. "I don't expect significant movement to the new integrated systems in the next 12 months," says Technology Business Research's Schreck. "Because it touches upon so many issues, it will be a longer-term transition, one that will take three to maybe five years."

SUBSCRIBE TO OUR NEWSLETTER
Stay informed! Sign up to get expert advice and insight delivered direct to your inbox

You May Also Like


More Insights