Open-source Lustre gets supercomputing nod

High-performance Lustre file system is getting its day in the sun as new start-up Whamcloud unwraps commercial offerings.


A new start-up called Whamcloud is coming out of stealth mode Wednesday with $10 million in private funding and a notion to disrupt the often academic world of supercomputing by leveraging the Lustre open-source project.

According to CEO Brent Gorda, the company is targeting the need for high-performance storage solutions based on the popular combination of Linux and Lustre for application and data storage environments. The company plans to offer support and services initially, with an eye toward a turnkey supercomputing setup with hardware and software components, in the future.

For those less familiar with supercomputing technologies, Lustre is a massively parallel, distributed, open-source file system used for large scale cluster computing. It's designed to enable I/O performance and scaling beyond the limits of traditional storage technology and potentially applicable to any enterprise storage environment where very high I/O bandwidth is required.

But it's a hard technology to master and has had stability problems in the past. Oracle owns Lustre, by way of the Sun acquisition, and is in the position to provide support but hasn't made any direct commercialization attempts.

It's a bit hard to tell how much commercial demand there is for Lustre support and services--and in fact, this may be why Oracle doesn't advertise such services--but I suspect there are a number of other products that could be atop the base software, as well as applications that could take advantage of the technology.

Lustre boasts more than half of the Top 500 supercomputing installations and seven out of the top 10 in 2009. It also set new I/O performance records when Oakridge National Labs achieved 240 GB/s on its Lustre filesystem the same year. Clearly, there is something there, we'll just have to wait and see what Whamcloud has up their sleeve.

Featured Video

Why do so many of us still buy cars with off-road abilities?

Cities are full of cars like the Subaru XV that can drive off-road but will never see any challenging terrain. What drives us to buy cars with these abilities when we don't really need them most of the time?

by Drew Stearne