Cognitive computing, which aims to simulate human thought and reasoning, could be considered the ultimate goal of information technology and is set to begin a new era in 2017, according to Kinetica’s Eric Mizell, vice president, global solutions engineering.
“The foundation for affordable cognitive computing already exists based on steady advances in CPU, memory, storage and networking technologies,” said Mizell. “A major breakthrough came with the advent of the Graphical Processing Unit (GPU) and its use in data analytics.”
There are four major trends that are driving the adoption of GPU-accelerated databases in this new 2017 Cognitive Computing era, according to Mizell. These include:
Graphical Processing Units (GPUs) deliver up to 100-times better performance compared to leading in-memory and analytical databases. This is due to the GPU’s parallelized processing architecture that contains over 4,000 cores, compared to 16-32 cores in today’s typical multi-core CPUs. The GPU’s small cores are able to efficiently perform similar instructions in parallel. This ability makes them ideally suited for these new compute-intensive workloads that are essential for real-time analysis of large streaming data sets.
Trend #2: GPUs will transform the Cloud into a “turbo-charged” network
Amazon already offers GPU-powered instances within its cloud service, and Microsoft and Google are not far behind, as they have both announced plans to offer GPU instances in their cloud offerings.