Univa and Grid 2.0

In his first discussion with the GCJ since taking the reins as CEO of Univa, Mike Ellis explains that Grid isn't just about reducing IT operations cost / complexity -- it's about enabling competitive advantage and time to market with new innovation.

GCJ: Univa has the phrase "Grid 2.0: Business Value Delivered" up on your web site. Start by telling us a little bit about the company's vision for the next generation of Grid innovation.

Ellis: As we all know, Grid has historically focused on compute problems, and tying together CPU cycles to throw against an intense workload or computation. But today, Univa and the Grid community at large are working to deliver Grid functionality that addresses some other common issues that are much more challenging and interesting, and that are meaningful across all industries - from biomedical and pharmaceutical, to industrial and high-tech manufacturing.

"Grid 2.0" is our way of expressing the broader capabilities that Grid technology provides, and it also describes opportunities where organizations can derive value from Grid beyond merely accelerating compute intensive applications.

Large enterprises have an IT environment that is extraordinarily complex in terms of what's needed to support the business-and it's becoming even more complex and harder to manage as many of these enterprises grow through mergers and acquisitions. They have many different strategies that they've executed against: at the transactional level, at the design level, at the product go-to-market level, and at the finance and IT level. Today, these large enterprises are looking at Grid as the road to a truly open and agile platform for their IT infrastructure. It's a common business issue that there is so much proprietary technology deployed throughout the enterprise. That environment can be at odds with the direction in which business is headed today, with more of a focus on open collaboration, global environments... and not just collaborating within the organization, but with their supplier base and partners as well.

The Grid can play a huge role in connecting the dots within and between organizations to tie together the various resources and to provide the management that's required. It allows them to really bridge the gap in this complex environment and create a collaborative view of computing that's aligned with the needs of the business. Univa is about providing a technology platform and solutions built on the open source Globus software that deliver the capabilities for enabling this new level of collaboration and reducing time to market for products and services in these enterprises.

GCJ: You come from a sales background. Selling enterprise software is known to be a tough grind in general. So we'd be interested in your thoughts on the challenge of selling a Grid solution, which presumably has a number of architectural implications above and beyond run-of-the-mill enterprise software sales, and would seemingly be an even tougher sale?

Ellis: Good question. From my perspective, companies are quite aggressively interested in building and providing an open platform for this collaborative world that I just described, both from a resource management level as well as a data management level. And they have come to the realization that the only way they're going to be able to solve this is with some level of open solution. So what we're seeing is that the folks that are selling a closed, proprietary solution today are the vendors that have to climb an increasingly steep hill with respect to their go-to-market strategy.

Companies can't solve the collaboration challenges that they have today on an enterprise scale-either on the resource management side or on the data management side-by providing yet another proprietary level of software to bridge the gap. They need some level of open platform... and that message is resonating for us really well within multiple industries.

Univa has some specific advantages that we believe uniquely position us to shepherd Grid into the enterprise as that open platform.

Our three founders obviously have very deep roots in an open source, open standards approach to IT. And Univa as a company has driven many of those best practices into the software components that we are building on top of the Univa Globus Enterprise, our commercially supported open source software product. There is a distributed data management capability, and an information management capability that we're building on top of Globus, as well as what we call service delivery management capabilities for enabling the automated, dynamic creation and management of collaborative environments. And we believe that through our heritage in the open source community, we can bring some of the best talent to bear on solving these issues.

At the same time, a big part of our corporate mantra is to continue to feed technology back into Globus and to the open source community. We believe that the stronger Globus becomes in terms of commercial impact, the better it is for Univa and the Grid community at large.

GCJ: One of the other sales angles that Univa seems to be pushing is that Grid isn't just about reducing IT operations cost / complexity - it's about competitive advantage and time to market with new innovation.

Ellis: Absolutely. In fact, it's one of the most compelling things about what Univa's doing. Again, it's not so much about compute power, it's about overcoming the whole series of constraints, resources, management and data issues that need to be addressed to enable the enterprise to accelerate the innovation process.

My background includes a number of years working with supply chain management technologies, and in that respect, I think there's a similar value for Grid in reducing time to market.

Large businesses are focused on what dials they can turn, both from a logistics standpoint, a staging perspective, a manufacturing perspective, a planning perspective-and managing all of the specific constraints and buffers that go into that plan, both within the enterprise as well as across all of their suppliers, distribution points, etc. What that adds up to is a very large collaboration problem.

Univa is focused on putting Grid to use in driving time to market reductions for large corporations, in a collaborative context, both around data management and product innovation. One of the key things that we're finding across industries is that companies are struggling to provide a really responsive environment that brings together data management resources as well as computing power, and makes them available to the knowledge workers that are actually driving either new product development or time to market improvements for current products.

What Univa's solutions provide is an environment where enterprises can accelerate that innovation through better management of resources, across compute, scheduling, data management and security, and by using our open platform as the underpinnings of the technology to accomplish that. You're talking about billions of dollars in terms of potential impact to these enterprises.

GCJ: So when you're selling into a large enterprise to speed up their innovation cycle, are you selling to the core enterprise data center guys, or are you selling into the R&D guys? Do you have to influence the guys that run the core data center, or is it a kind of peripheral group that you're trying to sell to?

Ellis: Actually, it's both. You need to focus on the business side of the house to demonstrate potential business value, and you need to get validation from the IT side of the house-the CTO and CIO-about the resilience of your technology, the capability of the platform, and you ability to deliver.

Our story resonates initially on the business side, because that's where we're trying to drive value in innovation processes and time to market. And that's where the impact will be felt on a company in a very broad way. For example, if you look at what's going on within the aerospace industry or the semiconductor industry today, billions of dollars of revenue potential are tied to the opportunity for reductions in the time-to-market cycle. There are senior executives within a broad range of industries that are trying to identify innovative things they can do to impact time to market. So this message resonates really well with them. They're looking for open answers to providing a collaborative environment that speeds up this process.

Now, at the same time, the IT side of the house is equally important to us, because they are very interested in new technologies that help them deliver against these same pressures. We're finding that the CIO/CTO is very interested in understanding how new methodologies and technologies can be applied because they're feeling the same business pressure from the COO and the CEO of these large enterprises - how do I compress this product or service delivery cycle?

In the '90s and early 2000s, many companies were applying enterprise resource planning solutions and applying supply chain management applications to solve their enterprise transaction management problems... essentially installing the basic, and necessary, plumbing for handling transaction-based activities faster and at a deeper level of execution.

Today what we're seeing in most enterprises is that they have those basic capabilities in place. What they're looking at now is how to address the needs of knowledge workers, the innovation leaders in these companies. They're trying to figure out how to provide an environment that speeds up their research and design process, that accelerates their time to market, and significantly improves innovation delivery and execution. It's worth billions of dollars in value to the enterprise. Univa is all about providing open, enabling solutions for these environments that drives time to market compression and innovation collaboration.

close window