Global Sources
EE Times-Asia
Stay in touch with EE Times Asia
EE Times-Asia > EDA/IP

Can a chip guy get Cadence moving?

Posted: 16 Feb 2005 ?? ?Print Version ?Bookmark and Share

Fister: Maybe there's something new to be learned from an old hardware guy or from a guy who has been a customer and knows what customers need.

The marketing department of Cadence Design Systems Inc. took a gamble by letting Mike Fister participate in the Design Automation Conference's CEO panel last June, only a month after he had assumed the presidency and post of CEO at Cadence.

Seated alongside veteran EDA CEOs Aart de Geus and Wally Rhines, of Synopsys and Mentor Graphics, respectively, Fister appeared to struggle on a number of topics that were admittedly arcane for anyone new to the EDA industry. But the gamble seemed to pay off nevertheless, as the former Intel Corp. executive presented himself to a room packed with EDA customers, executives, venture capitalists and bankers as a technologist and marketer with a track record of success.

Months later, Fister still does not appear to be well-versed on the more esoteric details of the EDA business. He does not talk about SystemVerilog vs. SystemC or assertion-based verification, for example. But he has become familiar with his company, his customers and his new responsibilities as the head of a public company that also happens to be the largest in the EDA industry. That makes Fister a key player in determining the health and growth potential of an industry on which the larger semiconductor business and, ultimately, the electronics industry rely.

On many occasions and from many sources both inside and out, the EDA industry has been accused of being in a rut, of refusing to grow up and expand its market potential, thus failing to clearly realize its true value for investors and Wall Street. To counter that crisis of confidence, Cadence believes Fister brings a new level of professionalism and a bigger, technical vision to the company and, perhaps, to the EDA industry overall.

Fister received a B.S.E.E. and M.S.E.E. from the University of Cincinnati in 1977 and 1978, respectively. He then went to work for Cincinnati Milacron Inc. and was a founding member of a company called Machine Vision International, which was focused on computer vision for robotics and generic automation. In 1987 he moved on to Intel, where he helped design the company's first Ethernet chips and I/O controllers.

Fister said that successes in those early designs developed into management opportunities, and eventually he was promoted to general manager of the performance microprocessor group. There, he was responsible for mainstream Intel microprocessors!notably, the P6, Pentium Pro, Pentium II, Pentium III, Celeron, Xeon and Itanium lines!before moving on to the enterprise platforms group.

With a track record like that, Fister could have taken a CEO post at other, larger electronics companies. But he said he was drawn to Cadence because of its role in furthering the technology as the industry grapples with the grueling pace of Moore's Law.

Cadence certainly benefits from having an EE, and someone with a customer's point of view, at the helm. Former president and CEO Ray Bingham, who is now chairman of Cadence's board, has a strong business background but had to defer to others for advice on technology strategy and vision. That became a problem as the folks in the adviser roles rotated in and out of the company. The upshot, by many accounts, was that Cadence fell behind Synopsys and Avanti in the IC implementation space in the late 1990s and temporarily ceded the top spot in the EDA industry after the merger of those two rivals.

Earlier this year Cadence regained the No. 1 ranking, largely because it was able to integrate technologies it acquired from SPC, Plato Design, Get2Chip and Simplex before the license renewal cycle kicked in during Q3. That rebuilding of Cadence's implementation flow was more or less completed just as Fister joined the company, but Cadence employees say his arrival furthered the momentum.

Food fight

Fister has earned a reputation as a motivator and a visionary. By his own admission Fister was more involved in the "long-term vision, rather than the details of how to solve a problem" at Intel and has relied on competent people to help make those visions a reality.

EDA veterans often say that former Cadence CEO Joe Costello was the last real industry visionary. Fister said he has not met Costello but is well aware of his legacy for motivating others and trying to grow the EDA industry.

Costello once likened the major players in EDA to dogs fighting over the same bowl of food, and worked to create more bowls!new avenues of growth!to feed everyone. Fister smiles when he hears that analogy and says he believes there is plenty of growth to be found simply by serving customers better.

"Maybe there's something new to be learned from an old hardware guy or, better yet, from a guy who has been a customer and knows what customers need," said Fister. "That's the legacy I'd like to leave!it's always being able to look at it from where the customer is."

As an industry leader, he said, Cadence has an opportunity to teach customers and others "a better appreciation for the foundational element of EDA." He compared the company's role with that of Intel, where "we were able to change the IT industry, with strong horizontal suppliers. Companies in the IT space used to have captive development resources and they don't do that anymore."

Fister said much of his success at Intel came from stepping back to look at an entire application problem and finding the entire solution. "It is taking 'learning' and turning that into experience and then turning that into product," said Fister, who believes the same method can be applied to EDA. Design automation companies can serve their customers by means of what he calls the "kit" concept!creating tool flows to address the design needs of complete application domains.

"What we do is reach across our technology potpourri and build kits that appeal to problems specific to that domain," said Fister. "It's less about 'can you synthesize the chip, lay it out and sign it off.' It's more 'can you architect the thing, work on it with a disparate workforce, how do you deal with software, how do you lay it out, verify it and prepare the chip for integration on the package and PCB?' That's what we mean by looking at the kit."

Also key to ensuring that customers have complete kits and that their application needs are met is collaboration with other vendors, he said, pointing out that the OpenAccess database facilitates this strategy. "I think we are broad enough where we can talk to a lot of those different customers and customize per application domain," he said. Fister said the most powerful customer relationships are partnerships, and he promised they would become a real emphasis at Cadence going forward.

- Michael Santarini

EE Times

Article Comments - Can a chip guy get Cadence moving?
*? You can enter [0] more charecters.
*Verify code:


Visit Asia Webinars to learn about the latest in technology and get practical design tips.

Back to Top