At 34 years old, Chris Kemp has a resume that many would call a career — Silicon Graphics systems engineer, Classmates.com chief architect, founder of online grocery technology company Netran and online travel company Escapia, CIO at NASA’s Ames Research Center and most recently the space agency’s first CTO for information technology.
“I’ve been incredibly fortunate to be given the opportunity to take risks,” he explains.
Now Kemp is back in the private sector, taking more risks and applying the lessons he’s learned along the way. And he’s still not shy about sharing his opinions, as GeekWire learned in a recent conversation with him at the Seattle offices of his new technology startup.
Kemp is the CEO and co-founder of Nebula, a company developing an IT appliance that aims to make it easier and cheaper for companies to run “private clouds” — letting them scale up or down quickly by giving commodity servers behind the firewall an infrastructure more typical of public clouds from Amazon, Google and others, at a lower price than paying those providers by the bit.
Nebula faces no shortage of competitors. Companies including Cisco, EMC and VMware have banded together to offer their own alternative, for example, but Kemp pans that approach as a “gold-plated, incredibly expensive, pathetic excuse for a cloud.”
Nebula’s appliances, he says, will “strike the appropriate balance between buying the cheapest commodity infrastructure yet providing a really incredibly reliable, scalable infrastructure service layer that sits on top of it.”
“What we’re trying to do is we’re trying to make it possible for CPUs, discs and memory — things that store bits, flip bits and move bits — to be 99 percent of the cost of your infrastructure,” he says. Nebula, he explains, is all about reducing the “commercial software and metal and crap you have to wrap around raw commodities.”
The company is an outgrowth of Kemp’s work at NASA and his role, at the agency, in the OpenStack cloud computing project.
Nebula currently has about 30 employees, and should easily grow to at least 50 over the next year, Kemp says. It’s currently conducting pilots of its appliance. It hasn’t announced pricing, but Kemp promises it will be “incredibly disruptively priced.”
The company is based in Palo Alto, with funding from some of the major players in the Silicon Valley investment community: Kleiner Perkins Caufield & Byers and Highland Capital Partners, along with Google’s first investors: Andy Bechtolsheim, David Cheriton and Ram Shriram.
Nebula also has a sizable presence in Seattle, and has recruited key employees from Amazon and Microsoft.
“I think I found the perfect recipe, which is to get a company funded down in the Valley and then put a large presence in Seattle,” Kemp said. “It’s Nebula’s secret weapon.”
Why does he prefer that approach?
“There is no comparison at all between the Seattle VC community, if you could call it that, and Silicon Valley,” Kemp says. “I’ve spent a lot of time trying to raise money here, and raised some money here. The relationships and the interactions I had with the VCs here are completely different, night and day, black and white, from the guys down in Silicon Valley.”
In what way? “It’s almost as if VCs in Seattle don’t want to take risk, at all,” he says. “They have to understand everything in a way that you’ve removed all risk from the equation before they invest.”
On the other hand, he says, “The tech community is phenomenal. There are a lot of really smart engineers here, and a lot of them came out of Boeing or had parents who came out of Boeing and did interesting things. … The University of Washington is a little like Stanford. They’ve got a fantastic CS program. I think the area attracts a lot of talent that leaves and does companies.”
The code name for Nebula was Fourth Paradigm Development, a nod to the work of Jim Gray, the Microsoft researcher who was lost at sea in 2007. Gray predicted that data-intensive science would become a fourth paradigm of scientific exploration.
Kemp explains: “I think every business over the next 10 years or so is going to reach a point where they really embrace data, and as they get to that point, they’re going to look very hard at how much they’re spending on their infrastructure. And when they get to that point, that’s when our phone rings. It’s expensive to do it the old-fashioned way.”
