Texas Data Center Summit: Hugh Miller Says Cloud Still Needs to Mature
by Josh Anderson
DALLAS, TX — Hugh Miller currently runs a CIO and Broadband consulting company and is serving as the CIO for Visit San Antonio (VSA), San Antonio Museum of Art, and Eva Longoria’s non-profit organization Eva’s Heroes. Hugh is also helping a few startups in the San Antonio area with his leadership and technical skills and has been consulting for communication and broadband providers. Hugh served as the CIO & CTO for the City of San Antonio from 2004 through 2016 and was responsible for technology leadership, innovation, transformation, and technology support throughout San Antonio. Hugh has also been honored as a Computerworld Premier 100 IT Leader for 2008, IT Leader of the Year in 2013, and SA Business Journals 2015 Tech Titan as Top Tech Executive. Tomorrow, Hugh will be a featured speaker at CapRE’s Texas Data Center Summit in Dallas, TX. In anticipation of the event, we spoke with Hugh about his career, his perspective on the state of Texas, and his thoughts on the latest developments in the data center arena (check out the first part of this Q&A posted yesterday).
CapRE: So let’s zoom out a bit. On a national or even global scale, what are you most excited about?
Miller: The fact that the ability to replicate what you are doing is becoming so much easier than ever before. The benefit that will have is that your disaster recovery components are going to be more affordable and viable. For a long time, those were just conceptual and theoretical benefits. To make them work was really, really complex, looking at timing and everything. But as we continue to increase the affordability and reliability of broadband circuits, and the latency reduction of them, you’re seeing that it’s a very viable and affordable way to ensure that what you are running for people stays in tact and accessible, even through very difficult situations, whether that’s some type of failure or a weather incident. I’m very optimistic about that.
CapRE: I think we all are!
Miller: I’m also excited about how you’re able to virtualize things – storage, servers, coupled together, where you can very seamlessly stand up what you want in a server and scale it up and down based on what you want. You can seamlessly configure a network based on what you’re needing and when you’re needing it. We can then bill you more on usage than the traditional model. Right now in Cloud data centers, it’s very similar to what you have in your physical data centers. You have to buy a server. Then you have to hope to get the most out of that investment. With the cloud side, it’s very similar. You can go in and scale out a specific categorical size of a server. Then, hope that it’s the right scale. But now you can look at the workloads, and move them up and/or down based on the usage.
So for organizations like Visit San Antonio, there is a pocket of time during the day that their financial system is going to get used a lot more than others. And think about weekends, holidays, etc, when they’re used a lot less. You should be able to move those resources up and down based on demand. It’s like energy companies. They monitor the use of energy and they can pour energy into certain business areas during the day, and then shift it over to the residential side at night. I’m looking forward to that.
CapRE: Ditto to that. Onto the flipside, though. What are you particularly concerned or worried about in our industry at the moment?
Miller: The maturity of the Cloud is still needing to grow. I think that the connectivity component is one that really has to be looked at – how do you scale it, how do you afford it? That’s very important to ask. Like a lot of companies, a traditional load system is going away. Pretty much everything in the next 5-10 years will all be IT encapsulated. With that is a quality of service need — that you scale your usage over that same time, in order of need. So like if I do voice technology, that has to be the highest priority of everything I put over my network. It’s the same with video. Because you’re putting it all on the same pipe. You’ve got to make sure that the type of circuit that you have and how it’s performing is managed well. That becomes your critical factor for everything. How you access servers, for voice and video, for conference, for example. There’s not a quick turn around if you need to change it. There are times it can take months to change a data circuit. That can be detrimental, especially if it’s not done right.
CapRE: Eek, let’s not think about that.
Miller: The other thing is just making stuff available. I mean the Cloud is growing. I get notifications every week bout a new component that’s becoming available. But then you want to consume it. A lot of these hosting companies like AWS and Google and Azure have these regional data centers. They’ll pop open a new feature, but it’s only available if your servers are in that data center. Then that becomes problematic. If you really need it, you have to move your servers. That’s a whole lot of risk and a whole process. So it’s very complex and very intricately woven. If you’re building hybrid data centers, it’s very complex. It’s very expensive if you don’t keep your eye on it.
Miller: Cloud computing is an arena of tech that everyone will consume (and in many ways, we all already are). To what scale depends on what organization you’re working with. So the need to really dig in and understand the underlying pieces of it is very important. Even if you’re rally knowledgeable and you’ve physically managed a data center, it’s very different to do so in a digital interface. It’s very important that you get that underlying understanding of how all of this is going to work. How to understand each different provider’s lingo. And then choose wisely how you venture into it. Contract wisely. Its not just a “go buy this and plug it in” type scenario anymore.
CapRE: You’re right about that Cloud statement – work aside, all of my personal assets are in the Cloud.
Miller: You know, with public cloud, you’re making yourself available outside of your building. Call it what you want, but that’s a Cloud service. It will continue to grow because there are a lot of things I can’t replicate very easily that a large data center can’t do. Even if you’re just running SAAS, it’s very difficult to replicate some of these things unless you’re a very large organization. Even then it doesn’t always make sense. More and more, and most people are already consuming it in their personal lives, we will continue to consume more and more. You just have to make sure you’re layering the components in a way that allows your business to function in a creative, innovative way. You can’t choke them out. But at the same time it’s got to be secure. And in many ways that can be an oxymoron.
CapRE: Ha, it sure it. Thanks for your time today, Hugh. We’ll see you in Dallas.