Industry Leader Q&A with IBM’s Bob Sutor

Bookmark and Share

A longtime advocate of Linux, Bob Sutor is vice president for Open Source and Linux at IBM. He recently talked with The IBM LinuxLine about the ways in which IBM has championed Linux over the past decade and the opportunities Linux is now enabling for customers.

LinuxLine: If an organization is considering using Linux, why is IBM the place to start?

Sutor: I think they would start with us by first looking at our experience. This isn't something that we just adopted very recently. This is something that we have been working at for a decade. Linux itself goes back more than a decade, but IBM was certainly one of the first commercial backers. You have to look at our commitment to Linux, a commitment to working with the community to make sure that Linux runs on our hardware, and then look at our commitment by supporting it with 500 of our software products on Linux. You can look at our commitment to providing productivity tools on Linux desktops. Lotus Notes runs there, Lotus Sametime and Lotus Symphony, which is a free productivity suite including word processing, presentation and spreadsheets, as well. If you look at what we have done with the 500-patent pledge in 2005 as well as many other pledges around open standards, IBM has done more than anyone else, more than any other corporate entity, I believe, to both contribute and to advance the cause of Linux and to demonstrate that business-critical products can run there.

It comes down to who has made the commitment, and who has made it an essential part of their business and who has demonstrated the innovation.  I don't think there is any comparison to what IBM has done.

LinuxLine: What does Linux offer to IBM?

Sutor: We think it is in many ways a way of optimizing things for customers. Linux gives us a lot more choices to really produce solutions for the customers that are tuned optimally for what they are trying to do in their business. 

LinuxLine: IBM has forged strong partnerships with Red Hat and Novell, key players in enterprise Linux and the community.

Sutor: That's right. In fact, Novell and Red Hat are both strategic partners for IBM. They happen be Linux, happen to be open source, but they are in that high level category of strategic corporate partners of ours.

LinuxLine: The recent news that Red Hat has joined the S&P 500 is being viewed as a sign of the maturity of Linux.

Sutor: I would like to congratulate Red Hat for being named to the S&P 500. That is quite an achievement and an extremely significant one, and the first for an open source company.

LinuxLine: How is Linux driving innovation? In what areas?

The first obvious one is operating systems. Linux is an operating system so people worry about things like file systems and they worry about security and performance and scalability and all of those things. Since Linux is open source and since it has so many people focused on this, we can really advance the state of the art there. But another aspect is the fact that many people just start with Linux in a fairly natural way and then build new products on top of it.

LinuxLine: Such as?

Sutor: Here is an example which is somewhat different from your typical business situation: Several months ago IBM announced that it was working on some artificial intelligence software that could play the game Jeopardy. When you think about it, and you think about the rules, it's a little strange because in order to answer, you have to come up with an appropriate question. And while I think many people pride themselves in doing this, it is an oddly at least so far human endeavor, to somehow go from a category and an answer to a question. It's not typically what computers do. We developed some research software to do this and it runs on Linux. Why? It's convenient. It runs on multiple operating systems. The same software could be moved from one piece of hardware to another piece of hardware. Linux does everything that people need it to do for an operating system. The software could run from a small device all the way up to a mainframe if we want it to-because it runs on Linux. This illustrates well, as people are trying to innovate, why Linux would be such a natural choice.

LinuxLine: Is there an example of how Linux has played a critical role in a real-world business situation?

Sutor: Linux has been used quite a bit in mainframe server consolidation situations. The Bank of Russia, for example, converted over from having hundreds of Unix-based servers spread throughout the company to consolidated mainframes, so that 60% of all the money that flows through Russia, flows through two IBM mainframes running Linux. With that, you get a lot more accountability because you have the possibility of reducing fraud. The data is now consolidated, not just the processing so you go from a situation which is very hard to manage to one where data is flowing through some very reliable, high speed computers and since you now have that information you can take it to the next stage and do analysis on that data. You can do the sorts of things for example that Cognos does. Linux has been a real key to running business-critical applications while saving our customers money and simplifying management in consolidation situations.

LinuxLine: Various distributions of Linux represented a 94% share of operating systems used by the supercomputers on the most recent TOP500 List. Is Linux unique in its ability to support demanding environments yet also work nicely for the desktop?

Sutor: I think one of the real advantages of Linux is that it is highly componentized. You can choose just enough, or you can choose more, and you can substitute in different parts of it, so the type of computer-the server infrastructure-that you might use on the fastest computers in the world is different from the configuration that you might use on your desktop. So you wouldn't necessarily need a word processor that happens to run on Linux on the desktop on one of these supercomputers. You might use different types of network access, memory access, operating system access on the supercomputers, though that technology may work itself down to the desktop over the next few years. 

I think that is a supportable statement, let me phrase it that way, that Linux is unique across the broad range of applications for operating systems for all these reasons. And it is done without fundamentally splintering. You might imagine that through the years as people came out with new types of computers or applications to use Linux you would have all these different forks and incompatible bits but that really has not happened. It is a testament to the management of Linux by people like Linus Torvalds and many other people that has kept this project on track, well-managed, making innovations and being delivered in a timely fashion, so that you can use it in so many different ways.

LinuxLine: Desktop Linux has been getting a fair amount of attention recently. Is it important because of the cost savings and functionality or does it represent more than that?

Sutor: We do find that for many users cost savings is extremely important for desktops, but I think it is important not to just say it's cheap, therefore I'll use it. People are looking for cost savings for very good reasons, but it is not always simple. People may save cost on the desktop operating system which allows them to spend that money elsewhere if they spend it at all. They may get more powerful hardware. They may get other types of service contracts. If they pay less for the operating system, they may be able to get more sophisticated applications that run on top of it, so it is important to look at costs within the larger economics of what you are trying to do there.

You might imagine a physical desktop or laptop, which looks vaguely similar to what you may have been running over the last few years, but there are also things like the Palm Pre, Android, which are Linux-based. A lot of the netbooks are running Linux, and the Google Chrome operating system, which was announced-we haven't really seen it yet, but it has gotten a fair amount of buzz-is based on the Linux kernel. What it is enabling is this innovation on the desktop. It is allowing people to go off and say ‘we can try it this way, we can do a small one, we can do a big one, we can ship it  with these types of applications to make it primarily web-based or we can make it something which is more traditional. We can deliver it all to your desktop or we can virtualize it so that the real processing is taking place on a powerful server somewhere else.' It is offering a lot more opportunity for people to get very clever to deliver exactly the right type of desktop, whatever that may be, to their employees that's tuned for what those employees do. Not everybody needs a full-blown heavy-duty desktop, but some people do. Linux allows you to deliver very light-weight desktops to very traditional desktops and everything in between.

LinuxLine: Does desktop Linux have a public relations impact in terms of allowing people to have a comfort level with Linux?

Sutor: I think you have to demo it. You need more high-profile people using it and showing that it works. That's what it boils down to.  I think part of the problem is this: People are used to some notion of one-operating-system-on-the-desktop-fits-all.  There is a dominant desktop operating system which frankly has been challenged by not just Linux but by Macintosh and by people doing more and more of their daily work in browsers and on smartphones as well. So this tradition of what is a desktop and what most people should have the same thing of, is just being attacked by innovation on all sides. It gives people the chance to see that other things work. 

We have spent a lot of time over the last few years working on things like the Open Document Format to produce the format based on XML which allows you to freely share your documents-your word processing documents, your spreadsheets, your presentations-from computer to computer, whether you are running Linux or the Mac and it is not tied to one particular product. With the great success of that, and it is being adopted in country after country around the world, it is allowing people to now look at Linux and free software that runs on top of that. We are breaking the old lock-ins and when those are broken, customers have more opportunities to see first-hand that the software really works and it does what they want.

LinuxLine: IBM recognized the potential of Linux early on and publicly pledged its support 10 years ago. How has IBM's commitment to Linux and the community been demonstrated?

Sutor: I think with an open source project, the first way of demonstrating commitment is to join the development community and contribute code. And IBM consistently, year after year, has been one of the top contributors to the Linux kernel.

But there are other types of communities as well when people talk about open source, and not everyone always remembers that. Yes, there are the software engineers that create the actual software that you are talking about but you have to ask: Are we supporters-do we help drive the messages around Linux, do we help advance it? That is almost kind of  a marketing approach to doing this-evangelism-and I think there it has been very clear as well that we have speaking of the value of Linux for 10 years to the press, to our customers, to analysts-in lots of different ways.

We have helped to separate out the different ideas about open source versus traditional proprietary or commercial software. We have described our hybrid business model. And, I think in many ways over the last decade we have helped advance the notions of what is a business model with open source itself. Over the last decade, IBM has participated by both creating Linux as a piece of software but then advancing it in the minds of our customers as being something that is extremely suitable for their critical business tasks.