SeaMicro generated a great deal of attention in the data center industry last year by flipping a basic strategy of most server makers. Instead of focusing on providing greater performance for general kinds of computing tasks, the Santa Clara, Calif.-based company tailored its machines to handle a narrower set of Internet-oriented chores. SeaMicro’s systems also conserve electrical power and space–two big issues for enterprises building out computer rooms.
The startup passed up Intel’s mainstream Xeon server microprocessor in favor of the tiny, relatively low-powered Atom chip—best known for use in netbooks—and packaged device for use in data centers. SeaMicro says its system, which packs 512 Atom chips into a box 17.5 inches tall, performs the work of 40 conventional Xeon servers while using one-fourth the power and one-fourth the space.
SeaMicro is the brainchild of Andrew Feldman, an entrepreneur with an extensive track record in both the venture capital and business sectors. Prior to co-founding SeaMicro, Feldman was an entrepreneur in residence at Crosslink Capital and US Venture Partners. Before that, he was vice president of marketing and product management at Force10 Networks, a data communications hardware manufacturer located in San Jose, Calif. Feldman holds an M.B.A. and B.A. from Stanford University.
I recently spoke with Feldman about the challenges associated with launching a computer company based on a radical new concept.
Seeking Opportunity
JE: Andrew, please describe your business and technology background.
AF: For the last 15 years I’ve been an executive in startups in Silicon Valley. My expertise is in leading programs for identifying where technology intersects with interesting markets. SeaMicro is my fifth startup. I’m not a technical visionary, I’m a person who sees market opportunities and identifies technical talent.
JE: How did you discover the new server technology?
AF: In the way these things usually happen in the Valley: you look. I saw the growing importance of power in the data center and I began to explore the cause of the power issue. As I began to understand the underlying roots of the power issue I started having breakfasts and lunches with different people in my network. We talked about starting a company that could address this issue. I talked to people who knew more about server technology than I did. I built a relationship network that was rich in expertise. I was introduced to our CTO, who had some ideas about how you could build a very different and much lower power server. So we began to flesh out those ideas and we began talking to customers. After talking to 30 or 40 customers we came to believe that we were onto something very interesting. With that, we decided to leave our day jobs and build a company.
JE: How did you find reliable advisors to move the company along?
AF: The Valley is filled with people who give advice; the key is reliable advice. What it means to be senior in the Valley is that you have a network rich in expertise. That expertise is not only in technology, but also comes as advice from customers. The customer base we chose to pursue in the data center was the same customer base that I would have been pursuing at Force10. So we have relationships with business leaders, technical decision-makers and visionaries, all of whom were generous with their time and were excited about our ideas. Listen to what your customers say and mean and you can convert that into innovation.
JE: How did you raise the capital necessary to launch the business?
AF: I have some experience in raising capital from the venture community. We put together a presentation that articulated what we thought we could build, how much it would cost and why we thought it was an advantage. We then made calls to the venture community and very quickly found interest. We were able to close our initial round very quickly.
JE: Was there any point at which you felt you had to compromise your vision?
AF: As a CEO one compromises many things every day. But looking back today on the past three and a half years, I can see that we built exactly what we set out to build. That doesn’t mean we’re not listening and taking in feedback constantly—and making modifications and including things—but the broader vision hasn’t been changed.
JE: Is promotion important to a startup’s success?
AF: Promotion at the beginning is not important at all, at least not for our type of business. We’re a hardware company and so the first 18 months were spent doing engineering. During that time we built an organization from three to 45 people and had only one person who was not a working engineer and that was me. During that first phase, for a complicated hardware project like a server system, promotion is not important all. Now, going to customers, listening to customers—what we call product marketing and management—is enormously important during that period because you’re folding that feedback right into your product. But public relations, advertising—we didn’t do any.
Once you have a product, you start the next phase of your company’s life. At that point people need to hear about you. While we don’t do any advertising, we do do public relations work and we participate in industry forums and gatherings of industry leaders and visionaries. They’ve recognized our insight and invite us to present frequently at industry forums. So we use a combination of word of mouth, press and participation in the most technical of forums to get our message to market. We’re not selling Diet Pepsi or burritos. Our customers are the largest data center owners and operators in the world. We know who they are. Our approach most of the time is to pick up the phone and call them.
Defining Entrepreneurship
JE: How would you define entrepreneurship?
AF: A passion for building companies, starting at zero. Those of us who do this every day know that the deck is stacked against us. Most venture-backed companies fail. Across history, David has fought Goliath many, many times and the vast majority of times David got squashed. But 4,000 years ago David beat Goliath and we’re still talking about that story.
JE: How does one begin to become an entrepreneur?
AF: There are personality traits that allow you to be an entrepreneur. You can be an entrepreneur in multiple dimensions—the world’s not just Silicon Valley venture-backed companies. But you need passion, a little bit of arrogance and the desire to be a risk taker.
JE: Can entrepreneurship be learned?
AF: You certainly can learn how to manage people in small teams, to hire first-rate people and to bring them together in order to rid them of the hassles of daily work and to allow them to completely focus on a problem. But you also have to learn how to tell your dad that you’re at a company he’s never heard of and that it may go out of business. Once people know the name of the company you’re working at, you’ve left the world of entrepreneurship.
JE: How does one develop the confidence to be an entrepreneur?
AF: One of the things Silicon Valley does really well is that it teaches us there’s no such thing as failing. That takes some of the risk away from entrepreneurship. You reach a point where you learn that you have a set of marketable skills and that makes you sufficiently confident. You know that if your effort blows up your family won’t be out on the street. After all, good engineers are hard to find.
JE: Are entrepreneurs control freaks?
AF: It’s the ultimate act of non-control, being in a tiny boat on the massive ocean of your market. There’s so much outside of my control in the server market. It’s a $40 billion market, but we’re a little company playing with the biggest of the big players. I have no control over the macro economy or business in Japan. There’s a huge number of things that impact every company that I have no control over. As an entrepreneur you live with tremendous uncertainty.
JE: What skills does an entrepreneur most need?
AF: A sense of humor and common sense, although I’m not sure how you learn these things. There are some people who are unbelievably smart, technically brilliant, yet on the common sense side they struggle a little bit. But that’s why you have a team. That’s what the organization does for you, it mixes technical brilliance with common sense and all sorts of other skills. The whole is greater than any of the parts.
JE: What’s a rule that can’t be broken?
AF: The coolest technology without customers is a dead business. One always has to remember that the goal at the end of the day is to build something that will be sold, that is what will drive your business decision making. It’s not about coolness or elegance; it’s about what customers want and what they’re willing to pay for.
JE: Who or what gives you inspiration?
AF: I’m always inspired by the underdog. The reason you’re playing basketball at Butler rather than at Duke is that Duke didn’t want you. When the underdog steps up and through tremendous teamwork, drive and hard work topples the big dog, that always ignites my passion.