Piston Cloud Computing co-founder Josh McKenty has laid out the same objective on several occasions, from speaking to attendees at the inaugural OpenStack design summit to addressing friends and colleagues at his company's launch party.
"Let's finish where we started."
Piston Cloud Computing, the San Francisco-based company McKenty co-founded last year, recently released the first private cloud operating system based on OpenStack, the open source infrastructure-as-a-service framework that was first designed by a group of researchers, McKenty included, at NASA. For a company that is less than 1 year old, McKenty is hoping its Piston Enterprise OS is the next step to fulfilling that goal.
LOOKING BACK: Ex-NASA cloud builder's startup boosts OpenStack software
McKenty says Piston Enterprise OS, otherwise known as PentOS, is designed to bring customers "private cloud that you can actually use." Effectively, as McKenty described it, Piston Cloud Computing hopes to make PentOS similar to Linux, but for private cloud management. The company's aim is to provide an alternative option for private cloud management, while maintaining security, reliability and navigability for the IT managers tasked with overseeing the network.
The origin of Piston Cloud Computing as a company tells a lot about what it intends to do with its new OS. The motivation for PentOS stems from McKenty's work with NASA's Nebula Cloud Computing Platform, where he served as the technical architect on the original compute components of OpenStack. Once he got past a bit of reluctance to take the job at NASA -- a result of his entrepreneurial spirit and the known limits of federal budgets -- he joined a group of researchers that would bring groundbreaking innovation in an unlikely setting.
"We're basically startup people and we did this impossible thing of running a startup inside the federal government, at kind of a period in time where that was possible in a certain amazing way," McKenty says. "There was a lot of new optimism about the role the government could play in technology innovation."
While working with NASA Nebula, McKenty says the startup mentality could only last so long. Operating under the umbrella of the federal government meant that budgets were set far in advance, and little else was flexible. From there, the researchers' work on OpenStack would reveal a potential that McKenty thought was being ignored.
"I didn't believe that OpenStack should be a chance to reinvent everything or change every part about how every business does IT because you can't get adoption there where it really matters," McKenty says. "You've got to be able to make incremental progress, and a lot of that is being able to tie into existing enterprise systems. There are all these examples around lost data, around authentication systems, and around cost visibility and accounting. These are boring things, but they're really important boring things. And I guess I have this perverse mindset where I can get excited about solving boring problems."
While working with NASA Nebula, McKenty recognized that the demand for actual products and services based on the OpenStack framework was high, and would only rise as more organizations gravitated toward the private cloud as the basis for their overall IT infrastructure.
"I wanted to make it something that everyone could use," McKenty says. "The biggest challenge we had at NASA Nebula was that once we announced it and started using it, we had demand from every federal agency, from foreign governments and from state and local agencies, and we weren't allowed, we didn't have a mandate to go and provide it as a service to them. They all wanted this cloud. They all wanted what became OpenStack."
In McKenty's view, this demand would be the same, if not higher, in additional markets. So when Rackspace acquired the team of researchers McKenty had been a part of at NASA Nebula, he had a choice to make: continue working on a foundation on which many companies may one day lay their foundation, or spin off and provide products that would help such companies do so. Tempted by the opportunities and potential he saw, he chose the latter.
"Really, I love the team at Rackspace," McKenty says. "They've done a brilliant job in building a community around this project. But they're not a product company. It's not really in their DNA, and I really felt that there were important things that needed to happen with OpenStack that were not going to happen except in a startup. And that was the genesis to launch Piston Cloud."
What remains to be seen is whether the reality of the private cloud market will meet McKenty's expectations for PentOS. According to Jay Lyman, senior analyst covering enterprise software for 451 Research, a demand for alternative tools for private cloud management and IaaS tools has arisen as more enterprise customers have gravitated toward the private cloud.
"There's a lot of interest and a lot of desire to at least take some small steps by enterprise organizations into something like OpenStack, [something] other than Amazon, other than VMware," Lyman says. "Sometimes it's a matter of cost, sometimes it's a matter of performance, but they have sometimes hit a wall in wanting to have an alternative."
The key to success for PentOS will be the level of support the company provides for its open source software.
"It's a pretty good match for what enterprises seem to want in terms of that marketplace, that diversity, that variety of choice in something like OpenStack, yet with the enterprise credibility, a phone number and the true higher-level technical support that they might need as an enterprise," Lyman says. "I think that PentOS is a fairly good match for that sort of demand, you know, kind of the best of both worlds."
A key aspect of this reliability lies in the foundation of Piston Cloud Computing and where it came from, according to Lyman. "Piston has some impressive leadership," he says.
Of course, the kind of impact this new product will have remains to be seen. With the kind of demand for new options in terms of private cloud tools, though, the outlook is optimistic.
"It's a growing market and we see continued demand and need for alternatives," Lyman says. "There's a lot of appreciation out there that OpenStack and the vendors involved with it exist."
Colin Neagle covers Microsoft security and network management for Network World. Keep up with his blog: Rated Critical, follow him on Twitter: @ntwrkwrldneagle. Colin's email is cneagle@nww.com.
Read more about software in Network World's Software section.