IT Management 2.0: Blame virtualization
Guest editorial by Bob Quillin, EMC
In 2008 -- somewhere in between the recessions, bailouts, elections, gas price insanity, Fannie Maes and Freddie Macs -- the "Data Center 2.0" vision reached critical mass. Personally, I blame it all on virtualization.
After speaking to hundreds of IT teams over the course of 2008, one thing's for certain: virtualization in all its forms - server, storage, and network - has taken root across the board. Granted, the depth and breadth of deployment varies from team to team, but virtualization changed the face of the data center in 2008. Given its cost benefits in terms of consolidation, IT managers have little choice but to drive virtualization hard and fast. That's the good news. The bad news? How do we manage all this stuff?
Welcome to the jungle In late 2008, the independent industry analyst firm Enterprise Strategy Group surveyed more than 150 IT managers in various stages of deploying virtualization technologies. The results? Only 24 percent were very confident their current virtualization management tools were sufficient to maintain existing service levels. That's a problem -- because in 2009 -- my guess is most of us are expected to improve service levels while reducing expenses at the same time. And while virtualization holds the promise of helping us accomplish these goals, the exact opposite seems to be happening.
While we're able to realize incredible cost savings from server and storage consolidation, it often costs more to manage virtualization because we're trying to manage it like we always have -- with the same tools. In many cases, we've created yet another IT silo to manage it, too. While virtualization is easy to manage in the lab or on its own, the data center is a jungle -- a diverse, complex ecosystem with hundreds of IT species and a delicate management balance. What's necessary is to introduce this non-native species (Virtualization Management) into the data center in a way that leverages its potential to change how we manage, while blending into IT's newest best practices, processes, and systems. This is the key challenge for 2009.
Talkin' 'bout my automation The answer for 2009 starts with another major IT wave that's now cresting: IT Service Management (ITSM). With unreliable IT service delivery, outages & interruptions, IT today can only viewed by the business as being inconsistent and unreliable. However, ITSM and ITIL process efficiency initiatives are able to bring a service-level focus to IT that transcends technologies - one that must be applied to the new data center and virtualization. ITIL addresses the biggest issues of configuration, change, incident, & problem management, but until we tie service desk and service catalogs to the virtual and physical infrastructure -- we'll be lucky to maintain current service levels let alone improve upon them. The second part of the answer comes from leveraging data center automation and compliance to eliminate error-prone manual processes. Virtualization and resource pooling should make this easier and -- as 80 percent of IT problems are caused by misconfigurations - clearly this is a requirement.
Finally, we need a way to tie this all together - to have a common view or model of IT that maps both physical and virtual worlds together. If you're not taking advantage of application dependency and discovery technologies that map across both P & V -- you're definitely managing the 2.0 data center with a 1.0 toolset. And with those discovery and dependency mapping tools, you're on your way to a real CMDB (configuration management database) able to track virtual and physical configuration items (CI's) and virtual dependencies, too. Tie that into your service desk and service catalog - now you're talking IT Management 2.0!
So, do you feel like I do? Maybe it's time to make a few new IT New Year's resolutions for 2009.
Bob Quillin is the Senior Director of Product Marketing for EMC's Resource Management Software. He can be reached at quillin_robert@emc.com.