Data Center Modernization Strategies

Industry Insights

Joe ClabbyData Center Modernization Strategies

A Q&A with Joe Clabby

Joe Clabby is President of Clabby Analytics, a technology research firm co-owned with his wife Jane. Joe covers a variety of IT products including analytics, system software, infrastructure management and security. The firm has covered products from companies including IBM, CA Technologies, Extrahop and New Relic.

 

hr-stratacloud-quote

StrataCloud: Do traditional data center managers and engineers like the changes going on now with cloud, DevOps, and software-defined infrastructure?

Clabby: There’s excitement around the cloud because it greatly improves the efficiency of IT, translates into lower operating costs and greater profits – and best of all, it breaks down organizational silos. People tend to resist change – but they’re not resisting the cloud anymore. They know they’ve got to move toward cloud architecture or they’ll be left behind running inefficient data centers with islands of information held in siloed departments.

As for the downside of clouds, enterprises need to invest in new tools and utilities to track and tune applications in the cloud (application performance management) and they have to understand and deploy a cloud security model. Further, IT managers have to move from managing physical systems, as they’ve been doing for decades, to managing virtual resources in the cloud. You’ve got to teach some old dogs some new tricks.

DevOps is the whole idea of cranking apps out rapidly using Agile methods. IT knows that apps are what drive business. The upside is, deploying more applications more quickly pleases internal users as well as customers. The downside is that mobile applications written using DevOps methods must be integrated with enterprise business process flows which takes a bit of work. Software-defined storage we write about a lot, but we haven’t seen the uptake that you would expect. I think that’s because there’s just too much else on the table right now in terms of change. It’s a matter of priorities.

 

StrataCloud: How do companies decide to go with a converged infrastructure solution or Best of Breed when designing new data centers or replacing old hardware?

Clabby: It’s about skills. Large companies will often have the skills to do the best of breed approach. Organizations that don’t have the expertise will go to converged infrastructure. Blue Cross Blue Shield of South Carolina is an example of a company that has adopted the DIY approach. The company has an approach to managing IT as a profit center where they support internal users but also sell solutions to other insurance firms. They have developed their own methods for developing applications, managing operations, business process workflow and so on. They can do this because they have 2000 people in IT. Managing Picasso is a book about their DIY approach. Right now, we’re seeing that converged infrastructure adoption is growing more in small and midsize companies that don’t want to or can’t do infrastructure design and management on their own.

 

StrataCloud: Any other decisions that will be critical in the coming year for IT leaders when it comes to data center modernization?

Clabby: Two months ago we were coming back from New England to South Carolina and stopped off at a business called Steel Orca. The company is building a monster data center utility business. It’s the most sophisticated data center I have ever seen in my life. The idea of utility computing is coming back. Computing should be sold like we buy power from electricity. You can plug in easily and pay for what you use and your connection is super-fast. The customer has to do absolutely nothing to keep things going. That’s not the same with the public cloud providers like AWS or Azure. Nobody’s really doing this utility computing model like Steel Orca right now.

 

StrataCloud: What are some of the key metrics IT leaders look to as a way to understand the performance and efficiency of their environments.

Clabby: I think less about specific metrics than about managing applications as they flow through the cloud and troubleshooting issues as they occur. That’s where companies like StrataCloud are spending their time by helping companies quickly deploy applications and perform capacity planning, monitoring to meet service-level objectives, rightsizing and self-healing. Companies care more about service levels than metrics generally, and that means supporting all the software to keep those service levels where they need to be.