By Julie DiBene
Director of Marketing
NVXL Technology, Inc.
The year is 2042 and it is 6 a.m. on a weekday. Music queues up, lights glow softly and that’s your wake-up call, your cue to get out of bed. Energy-conscious sensors turn everything off once your feet hit the floor and your shower starts cranking as you head into the bathroom. By the time you cross your bedroom floor, the water temp is exactly 107.5-deg F. As you step into the shower, you listen to your schedule for the day, a function you queued up ages ago because analysis has shown that you really do your best thinking in the shower.
Your Personal Assistant (affectionally known as PA), who also manages your health, informs that your last blood work panel showed a slightly elevated cholesterol level and, because genomic sequencing also showed a family history of high blood pressure and strokes, oatmeal it is. Also, much to your chagrin, you have gained three pounds so your PA schedules extra workout sessions for the foreseeable future. You programmed your PA, truth be told, to be a bit of a jerk.
Lunchtime is spent on the treadmill in the company gym. Once you start working out, your PA endears itself further by prompting you to work harder and elevate your heart rate.
Post workout, you video chat with your husband who works as a botanist improving the hydroponic gardens that are now on virtually every street in America, ensuring a continuous food supply. You both discuss the most recent reports on your daughter’s progress in piano which has been stellar. You insisted on music lessons because her genomics analysis showed a clear aptitude for music. The piano teacher, who is not human, continues to mix up lessons and approaches to learning because analysis also determined that this is the way your daughter learns best.
You end your day sipping a nice big glass of wine to unwind even though your PA likes to remind you that you have several relatives with addiction issues and a genetic predilection toward alcoholism. You don’t log your food and liquid intake in your PA precisely because you don’t want the insurance companies using it against you. The issue of health-related privacy is mired in the courts in 2042 so you side step the matter by electing to not record your intake.
This story you just read is closer than you think to becoming our daily reality. The technology is already in place in many cases. So, what’s stopping us all from having this life? In two words: compute performance. It takes huge amounts of compute power to process data the data in just one human genome.
To even begin tapping the collective potential of genomics and artificial intelligence, compute performance needs to be vastly accelerated. This can be achieved in a number of ways however, in order to make all these programs and applications part of everyday life, the cost also needs to come down dramatically. This is where NVXL is poised to change the landscape of compute performance.
NVXL is the brainchild behind a brand-new compute acceleration technology called polymorphic acceleration. This revolutionary approach addresses a multitude of compute intensive applications such as deep learning, genomics and transcoding. We are building a platform from the ground up to accelerate and scale compute performance. Coupling the power and flexibility of FPGAs, NVXL’s polymorphic acceleration platform assimilates seamlessly with deep learning frameworks and leverages a data flow architecture that utilizes dynamically provisioned instances crafted from a pool of acceleration devices.
Currently compute performance is accomplished in several ways, other than the traditional over-provisioned approach, time-based scaling is commonly used in modern cloud datacenters to provision computing resources based on historical workload patterns. This approach utilizes granular resource provisioning, but the overall compute resource utilization is still less than optimal. Right now, silos of specific computing architectures are the norm whereas polymorphic acceleration offers software-defined acceleration platforms in real time using a flexible accelerator pool that can easily be expanded and upgraded. Real-time scaling is designed to meet minimum workload demands. Using this approach, different image detection/recognition algorithms can be dynamically loaded, computing resources can be scaled up and provisioned on the fly so that the system is off doing other tasks until needed. This keeps the system utilization high, enables much higher levels of efficiency and reduces the Total Cost of Ownership (TCO). Imagine then, service providers enabling customers access to tens of elastic ExaFLOPs of computing power across 1000s of compute devices with a dramatically reduced TCO. Imagine a brave new world where massive leaps forward in technology and research can be accomplished because, for the first time, polymorphic acceleration will enable levels of compute performance never achieved before.
As high-performance computing in the cloud continues to evolve, deep learning, genomics and transcoding will all strive to move forward in the quest to solve highly complex problems. NVXL is leading the way, with a revolutionary polymorphic acceleration platform that enables disaggregated, heterogeneous compute acceleration with unpreceded scalability. Tune in for the next installment of this three-part blog, A (Future) Worse Day Ever in Your Life Made Better by Smart City Tech.
In the meantime, I invite you request a demonstration of this breakthrough technology or to continue the discussion of polymorphic acceleration and how we can help you take your compute intensive applications to the next level by contacting us at: Contact