Sequent was overmatched, CEO says

Market capitalization of $700 million just wasn't enough for the maker of high-end servers at the center of corporate operations.

Stephen Shankland Former Principal Writer
Stephen Shankland worked at CNET from 1998 to 2024 and wrote about processors, digital photography, AI, quantum computing, computer science, materials science, supercomputers, drones, browsers, 3D printing, USB, and new computing technology in general. He has a soft spot in his heart for standards groups and I/O interfaces. His first big scoop was about radioactive cat poop.
Expertise Processors | Semiconductors | Web browsers | Quantum computing | Supercomputers | AI | 3D printing | Drones | Computer science | Physics | Programming | Materials science | USB | UWB | Android | Digital photography | Science Credentials
  • Shankland covered the tech industry for more than 25 years and was a science writer for five years before that. He has deep expertise in microprocessors, digital photography, computer hardware and software, internet standards, web technology, and more.
Stephen Shankland
7 min read
Sequent, a company that designs high-end servers at the center of corporate operations, this week lost its battle to remain independent.

The company won acclaim for its products, but after more than a decade trying, it couldn't penetrate the top tier of computer manufacturers. IBM thus moved in to acquire Sequent for about $810 million.

For chief executive Casey Powell, the move was a bittersweet chapter in a story that began 16 years ago when he quit his job as general manager of Intel's microprocessor operations to start the Beaverton, Oregon, company. The takeover means Powell will lose his independence, but his technology now has a much stronger chance at reaching its potential.

Sequent's acquisition is another indication of the struggles faced by smaller computer companies with good technology in going up against the marketing muscle of the big boys, Powell said in an interview with CNET News.com.

While Sequent will be subsumed, its technology will propagate across the IBM product line, including its PowerPC-based RS/6000 servers, he said. Powell will remain in charge of Sequent products as a new division within IBM, reporting to Bob Stephenson, senior vice president of IBM's Server Group.

Project Monterey was the genesis of the acquisition, Powell said. Project Monterey is a collaboration between IBM, Sequent, and Santa Cruz Operation to write a version of the Unix operating system for Intel's next-generation 64-bit chips. Monterey will live on as a new operating system for PowerPC- and Intel-based systems with IBM using Sequent's NUMA (non-uniform memory access) architecture.

But the acquisition does raise some issues. For example, Sequent recently linked up with the FibreAlliance storage initiative sponsored by EMC, whose products compete with IBM's. IBM isn't a member and instead prefers to send standards through the Storage Networks Industry Association.

Powell acknowledges there are still issues to be ironed out. But overall, he's happy to have a second chance at fulfilling his company's dream.

CNET News.com: What's your overall reaction to being acquired by IBM?
Powell: I have mixed emotions. I've been here 16 years doing this. Our intent was always to stay independent, and I think everybody starts off that way, but I have to be a responsible individual and make sure I'm doing the best thing for our shareholders, for our employees. It's pretty clear the way that the way the consolidation is happening in the industry today, it's moving very fast. It's almost like musical chairs. You don't want to get left without a seat. If you look at who's managed to survive from companies that started off as minicomputer, there aren't very many left.

We feel that the technology that we have is significant, and obviously IBM does as well. If we're going to be part of another company, you might as well choose the biggest, best, and baddest one out there. We're feeling very positive about the fact that IBM has stepped up and selected us. It's pretty clear to me this represents an opportunity that may not come around again. With our technology and our people and our solutions, and IBM's channels and people, I think that we can really be a threat.

Q: Did you approach them or did they approach you?
A: We started to work together last October on a project called Project Monterey, which is to do an operating system. It evolved out of that. The more we worked with them, the more obvious it became to them that we had something that was important. And the more obvious it became to us that we could work within that environment.

Q: Do you see this news more as an affirmation that your technology was good or as an admission that your business wasn't good enough?
A: It's an affirmation. Here's what it comes down to, what [IBM's] Bob Stephenson said...With the combination of our technology and IBM's reach, this is going to be quite a threat in the industry.

It's very difficult when you're small, when you're a sub-billion-dollar company, playing against guys who are $20-, $40-, $60-, or $80-billion entities. You get out-imaged, out-talked. People don't hear you. You're just sort of not there. You can fight and fight and fight. But you have to be a certain size before you get the critical mass.

You get a company like Hewlett-Packard, these guys can spend more on corporate image advertising than we deliver in a year of profits. When you know about the game, you can go there, and you can give a good account of yourself and you can win.

It's hard to be the little guy on the block and have really great technology and get beaten, just because the other guy is big. Right now, we're going to be bigger. It's going to be fun for a change.

Q: You were talking about consolidation in the server industry. What other examples can you give of server companies that have gone by the wayside?
A: DEC's pretty prominent, Tandem's pretty prominent. If you look at the list of companies who were in the minicomputer business when we started our industry, Digital was second-largest computer company in the world. And then you had companies like Prime, Interdata, Data General, who's around today, but Data General today is predominantly a disk company. There are a whole variety of companies that are just gone now, that were either just absorbed by others or just went out of business.

Q: How will IBM use the Sequent products and technology?
A: The [Sequent] products they will sell. There's no question that with the amount of presence and reach that an $80 billion company has vs. an $800 million company, it's going to make a significant difference in our ability to approach and be effective in dealing with customers

The technology is going to find its way through IBM's entire product field. [Stephenson] referred to NUMA as a defining technology in the early 21st century. We will complement the NetFinity line, which is based on Intel, but I think you'll see a PowerPC-based product in fairly short order, and then also, where it's applicable, in AS/400 and System 390.

Q: How hard is it to bring the NUMA architecture from one chip to another, for example from Intel chips where it is now to PowerPC chips? How long would you estimate it takes to produce a PowerPC-based NUMA system?
A: It's not hard at all. You port the operating system and then you just rebuild the hardware with PowerPCs. It's not a big problem at all. Chipsets have to be redone. You could certainly have something you could demonstrate in a year.

Q: You chose to go with Intel chips, the high-volume product out there today. What would you gain from using PowerPC chips?
A: There's a large IBM installed base of PowerPC. IBM wants to make sure it continues to take care of its customers...If you're a computer company and you only have one solution, what happens is the customers's problems tend to get bent around that solution. If all you have is a hammer, everything looks like a nail. You can provide a solution as long as it's driving a nail into something. If it requires a screw or a bolt or a nut, then you need to have other tools. When you have a full bag of tools, then you can apply the appropriate solution.

Q: But you chose not to use PowerPC.
A: It's like any chip that tends to be lower volume. What we wanted as a company was an ability to get a chip that we could sell that we wouldn't have to sell against the guy we were getting it from. Buying [HP's] PA-RISC or [Sun's] Sparc, you're going to end up having to compete...downstream at a higher level of integration.

Q: Where have the NUMA systems been used?
A: Very large transaction processing with relational databases, and also data warehousing. We run
SAP [software] at Intel. We run all of Cisco--Cisco runs their company on our machines.

Q: Do you think NUMA architecture has the possibility to replace mainframes?
A: What's happening is the mainframe is starting to look more like servers, and the existing servers are starting to look more like mainframes. What you're looking at now is "recentralization," where those machines are bigger and bigger and bigger and they start to look functionally more like mainframes. The difference here is that you can do them so much more cheaply.

This stuff is getting bigger and bigger and bigger and more and more complex. You have to deal with complexity by centralization, because there aren't enough people anymore. So you need more and more people, [but] there are fewer and fewer available.

Q: What storage systems have you resold?
A: We team sell with EMC, we do it with Clariion, and we've done it with Hitachi Data Systems. We have a very strong relationship with EMC.

EMC is a customer of IBM, and IBM is a competitor of EMC. No one in this business is a stranger to the notion...of "cooptition." There are times when they compete and times when they cooperate.

Q: Do you expect Sequent systems to continue being sold with EMC and Clariion and HDS?
A: Absolutely.

Q: You just joined the FibreAlliance.
A: We're going to do business for the next 90 days. We have to run our business as an independent entity until we do the merger. There's every reason to go quickly, to get it over with, but you're right, it's going to create some awkward situations along the way.

Q: What still has to be done to complete the acquisition?
A: The first thing that has to be done is a notice has to be sent to shareholders. Then there's a shareholder vote. Concurrent with that will be preparing to move forward with government regulatory approval--all the antitrust stuff the government has to review to make sure everything's cool. That stuff will probably take somewhere in the area of about 90 days to do.