Sybase's CIO on Cloud Computing, Mobility

By Brian P. Watson  |  Posted 01-19-2010 Print Email
Jim Swartz, a longtime veteran of the CIO ranks, explains his view of the cloud’s pros and cons, as well as what he sees in store for IT leaders in 2010.

Sybase has made a major transition to become a mobility software provider. But beyond boosting the company's core offerings, CIO Jim Swartz is looking heavily at cloud computing, SaaS and maximizing the potential of the Millennial generation. 

Swartz, formerly an IT leader at several companies, including SRI International and SAIC, spoke recently with CIO Insight Editor in Chief Brian P. Watson about his priorities for 2010.


CIO Insight: What's your outlook for 2010?

Swartz: It's certainly been an exciting year because the company has had nine quarters of extended growth and profitability. So we haven't been hit by this downturn in the same way as other companies. We're all thankful for that and have continued to work toward making it happen.

We can't rest. It's a renewed effort. We have to think about how to posture ourselves to grow to that $2B mark and beyond--what kinds of information systems and technologies are people going to need, and how can we work with the business to improve their processes and advise them on the right technologies to improve their margins by doing things more efficiently and improve the top line to help bring more business into the company.

So what are your technological priorities for the year?

We as an organization are looking more and more into what workflow improvements will make us more effective and efficient. We're going to be looking at a new approach to on-premise versus off-premise application hosting. Not unlike many CIOs, I have a number of initiatives, with cloud computing right up at the top. What does it mean for us? What does it mean for the business? How can we leverage it? And how can we look at that cloud and push away all the foggy mist that exists in it?

When you stop and look at some of these metaphors, "cloud" is probably not the best word to apply to something that's trying to make you more effective and efficient.

We would love to go down the pipe and have the ability to use cloud services existing as a commodity and treat them very much like you'd use treat electricity or some other commodity service that you'd buy at the lowest price and move from vendor to vendor without disrupting your business.

We're probably quite a ways away from making that happen. I say that even as we've made some partnerships with cloud-providing companies, but unfortunately, the standards aren't quite there to make it real. In the next year, we're absolutely going to be looking at this more closely in terms of what it's going to mean, long-haul, because as an organization, to help the company keep its margins in perspective, I don't want to go out there and have to build expensive data centers. If I can use outside services, where appropriate, that's a great thing.

We've been looking at it with some skepticism. It's a no-brainer for SMBs, but bigger-company CIOs are weighing what it is, and whether it's an internal or external thing. What do you see as the big obstacles?

There is this issue around security, that you're going to be very cautious putting your financial information into a third-party's hands. However, some of the early ventures out into things like Salesforce, you are putting some sensitive information out there. And it's a multi-tenant offering, so some could see that as dicey.

But I think we're going to end up, in a period of time, with much more of a hybrid environment where you'll keep your sensitive applications on premise, but those applications will exist in what you will call your own internal cloud. You'll be able to use virtualization to move that information around in internal clouds. Instead of having a failover into another environment, you'll be able to push over into another site.

You create that competence to be able to do it, along with the security regimes around virtualized environments. As vendors create more standards to accept those virtualized data centers you build, then you'll feel more comfortable putting them out there. It'll take a while. There will be some successes and failures, as there are with anything. I still think we're five or six years off from the point where we'll send out something like an ERP into a hosted environment.

I'm just waiting for something to happen in the near-term where some SasS or cloud offering has some major failure. Then the business will have some trough of disillusionment, as Gartner would say, before people slowly and behind the scenes pick themselves up and improve their offerings. There's interesting buzz, but this is still five to six years away. Still, there will be some things that tend to work. It's interesting what Google is offering with Google Apps, and I see some people moving in a direction of they'll host their e-mail offerings off-site. It may make sense to small or medium business, but for a lot of us, mail hosting is just as sensitive as ERP hosting, because they're such sensitive information that goes out in attachments. So you have to think about that very carefully. We're looking at the effect of email re: attachments. We're trying to move attachments to a central repository so they don't go out with the e-mail and are protecting within the company. Those types of step-by-step protective measures--which are also cost-effective in terms of storage because an attachment doesn't go out 10,000 times, though a link might--can save a great deal. There are some of those incremental, less exciting savings areas that will come about in the next few years as people change the way they behave in terms of e-mail toolsets and take advantage of centralized computing and storing things more centrally in peripheral devices.

We use SharePoint inside the company, and it's become an fairly attractive tool. But we've also realized that as we put more information inside SharePoint, we become more tied in to that vendor. So we start to use more Office tools because they're so integrated. It makes us think about how tied in we are. Some of their products are very good, but there's a price for that. We are looking at open source and other alternatives to Microsoft as a primary vendor, but it's not without its issues or downfalls. Some open source office tools aren't as compatible with Microsoft as we'd like them to be. If everyone we're using them, it would be fine. But since most people are using Microsoft, it makes it harder to use alternatives to deal with customers. The kinds of things we can do with SharePoint and related toolsets, as well as collaborative tools like Facebook and Twitter and their corporate derivatives, we have to be doing more and more with those toolsets because people coming in are going to require it, and it's becoming more accepted across the industry for people to communicate over a Facebook-like toolset. We can't stop it. We can put dictates out saying we won't do it, but we'd have to back-pedal after some period of time. I'd rather lead the charge and shape its use, rather than being shaped by the demands of the marketplace altogether.

Some of these things are risky. If you want to get out ahead of it, you may have to take some risks and have some skunks-works going on inside to test this stuff out, so you have to be careful. And we've been caught more times than I'd like to remember trying to roll something out to early. So you have to have a balance in determining how and when you roll these things out so that you don't roll them out too early, but you don't want to be a laggard in rolling them out too late.

That's probably one of the biggest challenges IT leaders face in advising the business. Do stay behind and be a fast follower, or take the lead? Almost every decision you make has the potential to be a game-changing decision in the marketplace for your company, so it's very interesting.

The other thing is, the numbers are all over the map when it comes to things like cloud computing. It's supposed to be cheaper, but when you get under the covers, sometimes it's not as cheap. That makes the metrics and measurements more important.



 

Submit a Comment

Loading Comments...