What will the next generation data center look like? Why do CIOs often have a hard time getting an accurate or complete view of resources available in the data center? What solutions are available to do this? Get answers to these questions and more in this expert podcast.
During this podcast, Ben Grimes, Avocent chief technology officer and senior vice president of corporate strategy, discusses designing and managing next-generation data centers.
BIOGRAPHY: Grimes is responsible for serving the product development and direction of Avocent to best serve its thousands of enterprise customers. He also establishes the technical direction for the company by proactively evaluating companies and technologies pertinent to the future of the company and the industry. Prior to joining Avocent, Grimes had a 17-year career with IBM, most recently in the appointed position of distinguished engineer, an executive position held from 2002 to 2007. In that role, he served as lead architect for software development for IBM xSeries and blades. Grimes holds a bachelor's degree in electrical engineering from Purdue University.
He is also a featured speaker at the upcoming IT management conference, where he will talk about when IT and facilities merge: designing and managing next-gen data centers.
Read the full transcript from this video below:
Karen: Hello, my name is Karen Guglielmo, Executive Editor for searchcio.com. And I'd like to welcome you to today's expert podcast, featuring an interview with Ben Grimes, CTO and Executive Vice President of Avocent. Today, we'll be talking about next-generation data centers.
As the Avocent Chief Technology Officer and Executive Vice President of Corporate Strategy, Ben is responsible for servicing the product development and the direction of Avocent to best serve its thousands of enterprise customers. He also establishes a technical direction for the company, by proactively evaluating companies and technologies pertinent to the future of the company and the industry.
Prior to joining Avocent, Ben had a 17-year career with IBM. Most recently in the appointed position of distinguished engineer, an executive position held from 2002 to 2007. In that role, he served as lead architect for software development for IBM xSeries and Blades.
Ben holds a bachelor's degree in electrical engineering from Purdue University. Welcome, Ben.
Ben: Thank you, Karen; and thanks for taking some time with us today.
Karen: Great. And as I said, we're here today to talk with Ben Grimes about getting the most out of your next generation data center. I'll spend the next eight to ten minutes asking Ben to answer a number of questions about today's topic. So, let's get started.
First, I want to ask you. In your opinion, what will the next generation data center include?
Ben: Karen, it's going to include a lot of stuff. If you look at technology today compared to where we've been in the past, there is just a whole lot of new things that are coming into the world of the data center. We've had virtualization take hold and all of the things that go along with virtualization.
We've had advancements in CPUs. We've had advancements in storage and networking. We have new technology such as Cloud computing, which are changing the way that data centers are actually being architected. And what you're finding is that there is a whole lot of new things coming into the data centers that make them more and more complex to manage and control; because there's just so many new, different things, and that puts a whole lot of strain on the IT administrators and more emphasis on being able to simplify that complexity.
Being able to simply manage the environment that is changing in there because of all of the new and challenging things that are coming in, and those new and challenging things are creating newer type problems for the data center managers like things around power or capacity planning and managing those environments.
So, my belief is that as we put more and more stuff into these next-gen data centers, we have even more and more need for tighter controls and processes around how we manage the environments.
Karen: Okay. So, in your opinion, why is virtualization still such a fast-
Ben: A couple of reasons, you have the first one that came out when we started talking about virtualization; the ability to consolidate equipment. Being able to take advantage of the fact that, the servers or any of the compute devices in the data center world had excess capacity. With virtualization, you were able to take advantage of that capacity and actually then, start to get more value out of the compute environment that you had in the data center. That will always continue as we look at bringing in more and more powerful computers into the data center. You'll be able to virtualize more and stack more business workload on top of them; so, that trend will always continue.
But another interesting side effect of that trend is also around power. If we can take advantage of the compute nodes in the data center and stack more and more workload on top of them, we're able to actually look at things like being able to reduce power in the data center. Because I'm not in need of bringing in as much compute workload into the environment to do the same amount of work? So power will also be a side effect, benefit of virtualization as well.
Karen: So, why do CIOs and data center managers often have such a hard time getting an accurate or complete view of the resources that are available in the data center?
Ben: You know, Karen, it is a lot about the fact that there is so much stuff in there. We've looked at some of the studies that are out there that say, in a data center, sometimes the IT managers are just not even aware of about 30 percent of the equipment that's in the data center.
Because a lot of times, especially in those data centers where there isn't real rigor around change and configuration process, you could have business units like an engineering or a test lab bring equipment into the data center that they need in order to accomplish their job. But it may not actually have been known to the manager of that data center that that equipment has been brought in.
So, a lot of times when we talk to data center managers, they keep track of inventory or assets, literally, on a notepad or a notebook; possibly in some Excel spreadsheets or maybe Vizio files, but they don't have a good process or rigor around keeping those up to date. I mean, as soon as you complete an inventory, your accuracy begins to decline from that moment on because of the fact that others can bring equipment in or change equipment around.
So, not having a good process or rigor around change and configuration of the actual equipment in the data center causes us to have that hard time of getting that accurate or complete view of what they have in the data center.
Karen: So, what solutions are available for CIOs to get a better view or understand the assets and make better decisions about the efficiency of the data center?
Ben: I think there's two aspects to answering that question; there is the one around the process side. And as you look at processes like COBIT or the [inaudible at 00:06:11] processes, they talk a lot about putting really good change and configuration management processes in place for their environment.
The second aspect of it is having proper tools that will fit into those processes. And allow you to really manage those inventories. If you look at some of the things that are happening in the environment today, there are these asset management tools that are out there that allow you to bring your assets into a physical tool and be able to dynamically track them; tying them into a change and configuration process with a set of service desk or service management-type capabilities, so that you can put real rigor around what you door; an example, if you're going to onboard a new employee, you could have an automated process that goes out and sets up the appropriate active directory passwords and capabilities for that new employee it brings online into the data center. The appropriate images that that person may have need for access to.
It also allows you to look at things like how do you deploy to that person, their business workload and get them associated with e-mail and things like that? So, if you can automate those processes around different activities that affect the IT infrastructure in your data center. And then have proper tools that automate those processes, track those processes and give you a set of compliance capabilities. I think then, you'll have more control over what's going on in the environment.
Karen: Great. And finally, what are your best tips for IT in the facilities group best work together to optimize data center operations?
Ben: The key thing is that the two groups talk in the same language. If you look at the past, you've had folks that handled the facility side who were traditionally focused on things like power and cooling for the whole building. They talked in a certain set of engineering or technical terms.
And then you had a group in IT that was managing the IT equipment that talked in a different set of terminology. And the key thing is to be able to bring the two groups together at a table. And be able to talk in terms that they both understand.
Because what we find is if we can bring together both the facility side and the IT side and do appropriate planning for future needs in IT, you can start to lay out an architecture for both the facilities and for the IT side that meets both needs of both sides. What we've seen in the past and I've talked to several IT managers that they had built out new data centers say, three to five years ago. And literally, within one turn of the crank of the technology, the facilities were not capable of holding up to the new requirements from an IT perspective; from a power and cooling perspective, but if we could have brought those folks together prior to starting to build out the actual walls and starting to build out the actual power distribution in those environments. And talk about, what does the future look like in technology? What are the future needs from an IT perspective in terms of build-out [SP] and data center planning? We could do a better job of that up-front architecture before we start to build the walls.
So, there's tools out there that allow you to do better data center planning and modeling that can be used prior to actually laying the first brick for a building. That would then maximize the capabilities from a facility side and meet the needs of the IT side, both today and from a projection into the future.
Karen: Great. And on that note, that does conclude today's podcast. Thanks again to Ben Grimes for speaking with us today. Thank you all for listening and have a great day.
This was first published in December 2009