Sponsor TechView Atlanta and be seen by thousands of Atlanta area decision makers. For details, just drop us a line at sponsor@techviewatlanta.com
By Allan Maurer
The days of software developers catching a few winks at their desks while pulling 24 hour shifts did not go away when the Internet boom turned to bust nor in the latest economic downturn. The developers at AccelerEyes came into the company’s office across the street from the Georgia Tech Advanced Technology Development Center Saturday morning and didn’t leave until 7:30 a.m. Sunday morning, getting ready for the latest release of its Jacket Graphics Processor Unit engine.
Founded in June 2007, AccelerEyes is self-funded by its founders who came from its parent company, DivEyes. It has a dozen employees.
John Melonakos, AccelerEyes CEO and Founder
AccelerEyes won the 2008 Georgia Tech Business Plan competition and $50,000 in prizes over 29 other companies. Earlier this month it captured the GRA/TAG business launch competition top prize, winning $100,000 in cash and $200,000 in various business services, which CEO and founder John Melonakos says, “We’ll use.”
More than 70 companies submitted executive summaries to the GRA/TAG program, which were whittled to 40 or 50 full business plans, further culled to 13 semi-finalists and then to three finalists (Band Metrics and Talent Soup, each of which TechView Atlanta recently profiled, were the other two finalists).
AccelerEyes has developed a GPU engine that can run the MATLAB software applications 50 to 100 times faster on one of the newer GPUs than they run on regular CPUS. It also provides a graphics toolbox for the MATLAB. Scientists, engineers and financial analysts use MATLAB crunch the numbers in their massive data sets.
“People who care about speed and science and engineering code are the ones who care about our software,” Melonakos says.
MATLAB code is used by the aerospace and defense, automotive, biotech, pharma, and medical device, communications, electronics, financial services, industrial automation and semiconductor industries.
At lunch, you’ll commonly find company staff at The Tin Drum,
a Thai restaurant across from the ATDC.
“We started AccelerEyes because we realized there is a big niche for visual computing.” “When analyzing data such as DNA/RNA unfolding, a very complex process, MathWorks’ MATLAB users do not want to see data resolved as a numerical answer, but rather, want to observe the process visually,” says Melonakos. The same is true of other engineering and scientific disciplines.
Melonakos points out, that powerful graphics processing units (GPUs) driven by the video game market have or emerged or are emerging.
These have the potential to bring parallel computing to the PC, competing with less powerful CPUs.
Intel, for instance, has 1200 people working on its new Larrabee GPU expected to come out in 2009 or 2010. AMD is working on a GPU called Fusion. And Nvidia has developed a new GPU for speeding up complex computational code such as that used by MATLAB customers.
Most applications cannot even take advantage of the dual and quad core central processing units (CPUs) in PCs. These powerful GPUs, on the other hand, offer much broader parallel computing potentials of 256 or 1,000 cores.
AccelerEyes won the 2008 Georgia Tech Business Plan competition
and $50,000 in prizes over 29 other companies.
The main problem – one AccelerEyes solves for MATLAB users – is that the programming necessary for applications to take advantage of multi-threaded parallel computing is “very, very difficult,” explains Melonakos.
“To be able to use them, a programmer has to tell the application how to use or design into the application the ability to use those cores. The program constructs to do that are very difficult and the average programmer is unable to do it.”
“We enable MATLAB code to run on these emerging GPU chips,” he says.
Even though the programming task of taking advantage of 256 or 1,000 cores is even more daunting than dealing with dual and quad cores, the company’s Jacket product hides all that complexity from the MATLAB end user.
“Jacket enables MATLAB code to run on the PC as is, so they don’t have to do any weird multi-threading programming so their applications will run on these highly parallel chips.”
One benefit is that users get much higher speed visualizations, which is important for untangling genetic code data and other complex computational data sets.
After a seven-month beta phase, AccelerEyes started selling its first products in January to what Melonakos calls “a nice set of industries, 30 percent healthcare, 15 percent finance, another 15 percent in oil and gas industries and some smaller ones.” He says sales are going well.
Melonakos says those long developer hours are fueled in part by – what else? – caffeine.
“There’s a Starbucks right below us,” he says, “and we spend a lot of money there. A large chunk of our GRA/TAG winnings will probably go to Starbucks,” he jokes.
For lunch, he says the company staff favors The Tin Drum, a Thai restaurant also across from the ATDC. “We see entrepreneurial people there all the time,” he says.
Unlike some of those software shops of the heady Internet boom days, though, AccelerEyes doesn’t have any games running when they’re not coding, which is a bit ironic, because, Melonakos says, “We’ve got the best graphic processing units in the world here.”
Online: www.accelereyes.com
© 2009, TechView Atlanta. All rights reserved.
