Amit Vij on GPU-accelerated analytics databases

The convergence of big data, artificial intelligence, and business intelligence

By Jeff Bleiel
October 31, 2017
Dome ceiling Dome ceiling (source: Pxhere public domain)

In this episode of the O’Reilly Podcast, I speak with Amit Vij, CEO and co-founder of Kinetica, a company that has developed an analytics database that uses graphics processing units (GPUs). We talk about how organizations are using GPU-accelerated databases to converge artificial intelligence (AI) and business intelligence (BI) on a single platform.

Learn faster. Dig deeper. See farther.

Join the O'Reilly online learning platform. Get a free trial today and find answers on the fly, or master something new and useful.

Learn more

Discussion points:

  • The benefits of converging AI and BI in a single system: “You are orders of magnitude faster, and you have the ability to operate on real-time data, as opposed to operating on yesterday’s data,” Vij says.
  • The processing speed of GPUs: “The GPU really leverages parallel processing so you can maximize your throughput and take advantage of the advancements in hardware that have come about,” he says.
  • How GPU databases break down the walls between the data science and business domains: “Nowadays machine learning scientists and mathematicians can, in just three lines through SQL, execute their algorithms directly on data sets that are billions of objects,” Vij says.
  • How GPU databases integrate with machine learning tools such as TensorFlow, and how GPU applications can use the cloud

Post topics: Data science