Kaspersky Lab Utilizes NVIDIA’s CUDA for Faster Threat Identification

I’ve said this for quite a while, and still stick to it, that where GPGPU (general purpose GPU) is concerned, things seem to be slow to catch on. We’ve seen real benefits of the technology in the past, as the hugely parallel design of a graphics processor can storm through complex computations with ease. In many cases, the process can be accomplished much faster on a GPU than a CPU.

But up to now, most of what we’ve seen on the consumer side has been related to video encoding or enhancement and minor image manipulation, and most often, even those two are limited in what they allow you to do. But, there has also been some other interesting uses, such as cracking passwords. So with that, it led me to believe that we could see other similar uses for the home user. Not with cracking, but at least the ability to speed up things like virus scanners, malware-detectors and so forth.

According to a freshly-issued release from Russian-based Kaspersky Lab, it has begun using NVIDIA Tesla GPUs in order to speed up the process of detecting new virus’ quicker, which involves identifying unknown files, and then scanning through it to detect what kind of virus it could contain, or what kind of threat it offers. Compared to an Intel Core 2 Duo 2.6GHz, Kaspersky Lab says that the process completes up to 360 times faster on the Tesla GPU. That’s impressive.

As it stands today, this technology benefits Kaspersky Lab internally, not to the end-consumer, but as it has seen such stark increases in its lab, so much so as to actually adopt the technology, then it seems likely that the possibility of seeing it spread over to the consumer side for even faster scanning could be beneficial as well. Of course, the largest bottleneck on the consumer side is storage, not the processing power, but with the advent of SSDs, taking advantage of GPGPU could likely make a noticeable difference in a lot of scenarios.

The use of Tesla S1070 by the similarity-defining services has significantly boosted the rate of identification of unknown files, thus making for a quicker response to new threats and providing users with even faster and more complete protection. During internal testing, the Tesla S1070 demonstrated a 360-fold increase in the speed of the similarity-defining algorithm when compared to the popular Intel Core 2 Duo central processor running at a clock speed of 2.6 GHz.