X

Sun balances acquisition, innovation

Sun Microsystems tries to make clear at its two-day annual analyst conference that innovation pays--yet much of the company's biggest plans today came from innovation outside Sun.

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
4 min read
SAN FRANCISCO--Sun Microsystems tried to make clear one message at its two-day annual analyst conference here this week: Innovation pays. Yet much of the company's biggest plans today came from innovation outside Sun.

Two prominent plans Sun touted this week--the high-performance "throughput computing" plan and the N1 software to pool many servers into a single large computing resource--have roots outside the company. And in the future, Sun plans to continue to use acquisitions to bolster its own products.

"To give Sun credit, I think Sun does have a track record of a lot of in-house innovation. They have one of the more stellar lineups of chief technology types," said Summit Strategies analyst Dwight Davis. "But Sun doesn't have a lock on innovation. For some of the more high-profile initiatives under way, they have had to go outside."

Sun has budgeted about $2 billion for its research and development this fiscal year--roughly 15 percent of the company's expected total revenue--in an effort to ensure the company keeps its edge over competitors and its clout with customers.

"Our basic message here is that innovation pays," said Chief Executive Scott McNealy in a speech Monday to analysts.

Less emphasis was placed on another remark, though, by Mark Tolliver, executive vice president of marketing and strategy, who noted past successful acquisitions and told analysts to expect more. "There are some terrific M&A (mergers and acquisitions) opportunities out there. We will continue to do that in the coming year," Tolliver said.

Sun's archrival Microsoft has often been criticized for relying on acquisitions to expand its technology portfolio, and Microsoft hasn't raised much of a fuss about the issue. "The acquisition model rests more comfortably on them," said Illuminata analyst Jonathan Eunice. Sun, by contrast, "wants to project themselves as the innovators," he said.

Buying technology
Sun's reliance on outside technology is visible in Sun's throughput computing plan, which the company plans to describe in more detail in April. The promise of throughput computing rests on new server processors Sun acquired from Afara Websystems that can run as many as 32 jobs simultaneously.

But the processors aren't sufficient. Throughput computing also requires technology in which Sun has spent more than a decade improving: its Solaris operating system. Solaris, which still commands about $250 million of Sun's research budget, can accommodate vast amounts of memory and can juggle more than 200 tasks running at the same time.

Sun had been trying to build the throughput computing foundation with its own MAJC chip, David Yen, executive vice president of processors at Sun, said in an interview. There was one major shortcoming of that chip compared to what Afara accomplished, though, he said: "It was not Sparc," so running Solaris was impossible without major changes to the software--changes Sun pursued for a time.

N1, too, needed a boost from outside Sun. At the core of the technology is software Sun got through its purchase of Terraspring, and one key task, "virtualization" so computers don't need to know which physical storage system is storing their data, came from Pirus Networks.

Buying outside technology makes sense in many cases. Sun built its own clustering software, which lets one computer take over when a comrade crashes, but IBM chose to acquire technology for the task. The result: "IBM beat Sun to market by five years," Eunice said.

Over the years, Sun has acquired Clustra Systems for technology to keep server software running despite individual computer failure; LSC Software for the SAM-FS and QFS file systems; Star Division for its StarOffice software suite; Gridware for low-budget supercomputing work; the Trans division of Critical Path for running some mainframe software on Sun servers; Infrasearch for searching for information on peer-to-peer networks of computers; Forte for programming tools; and NetDynamics for running Java server programs.

Some of Sun's acquisitions, though, were duds. McNealy, referring to Sun's 1997 purchase of set-top box maker Diba, said on Tuesday, "About four minutes into the acquisition we said, 'Whoops!'"

New directions?
There are new directions Sun could take with acquisitions--particularly with software.

Buying a company such as Troux Technologies would get Sun software that could help it stitch together software running on different servers within an N1 collection, Eunice said.

And Sun could use database software, Eunice said, particularly as it embarks on its Orion plan to tie together many of its software packages with Solaris. Sun has close ties with database giant Oracle and frequently touts the lower-end MySQL, but those products are outside Sun's control.

Sun could acquire an established database such as Sybase or Borland's Interbase, but such a move would greatly displease Oracle, Eunice said.