Backstage with Johna Till Johnson

Network World later this month launches its second Technology Tour on the topic, Masterminding the New Data Center. Johna Till Johnson, founder and chief research officer at Nemertes Research and a Network World columnist, demonstrates in her keynote add

Network World this week launches its second Technology Tour on the topic, Masterminding the New Data Center. Johna Till Johnson, founder and chief research officer at Nemertes Research and a Network World columnist, demonstrates in her keynote address how the concept of the new data center is taking off in the IT community and shares the best practices IT managers are adopting. She recently discussed her thoughts on why the data center is evolving so dramatically with Events Editor Sandra Gittlen.

There’s a lot of confusion about what exactly is meant by the new data center. Perhaps understanding the shortcomings of the old model would help make clear the need for a new one.

When the economy came to a screeching halt, companies took a long hard look at what resources they had [in the data center], where those resources were deployed and what they were really costing. And they recognized a few things.

The average server and storage utilization ran around 10% to 25% – they had four to 10 times as much hardware as they actually needed. Storage requirements had been decoupled from server requirements – you no longer had to buy a server every time you needed more storage. And network gear – switches and routers – was taking up a much greater percentage of the data center footprint than before.

Companies also realized that while the overall trend had been toward putting computers closer to users, the advent of networking and IP meant that was no longer a requirement. Most employees are no longer at the corporate headquarters because they no longer have to be (87% of folks are at remote sites, according to Nemertes Research, and the trend’s increasing).

And finally, the people with the skills and knowledge to administer the corporate data center are often no longer the ‘mainframe guys,’ but folks with a background in networking, servers, storage or some other area of technology. A lot of the existing equipment wasn’t standardized and therefore was more expensive to maintain.

So what is the upshot for the ‘new data center’?

Data centers now can be located somewhere with low real estate costs and high reliability. They include a massive amount of networking, storage and voice communications gear as well as computers. To reduce overall cost of ownership, IT executives are increasingly relying far more heavily on standardized architectures (blade servers, clusters).

Data centers now emphasize highly redundant/reliable architectures. In the old days, tripping over a cable might have taken down an office. Now tripping over a cable could take down the entire Asia-Pacific region. Today’s data center designs increasingly involve automated management and application provisioning (you want to increase the utilization of resources while reducing the number of humans administering said resources). They feature technologies that lower operational costs and increase utilization, like virtualization and grid computing. And they are operated and managed by IT executives who are not necessarily mainframe/applications specialists.

What has your research shown about this shift? Are IT managers grasping what needs to be done? Or are tight budgets and lean staffing standing in the way?

By and large, IT execs ‘get it.’ Tight budgets and lean staffing are part of the reason for consolidation, not a barrier to it. And virtualization, grid computing and utility/on-demand computing aren’t cool new ways to jazz up your infrastructure – they’re practical approaches to reducing operational costs and improving performance.

What are the top five things an IT manager should do to make a move toward a new data center infrastructure?

1. Standardize on server architectures.

2. De-couple servers and storage. Appoint a ‘storage czar’ to oversee how storage is delivered.

3. Take ownership of facilities. Review your HVAC requirements in light of the network, storage and communications gear you’ll need to support (keep in mind that most data centers assume that computing platforms comprise the major tenants; that’s no longer the case).

4. Invest in redundancy – networking and platform.

5. Explore next-generation hardware and software: grid computing, high-availability switching, storage virtualization.

How can IT managers sell this concept to the boardroom? What’s the No. 1 benefit they should point to to gain backing?

Lower operational cost. You save money and improve performance by automating operations.