A new partnership with Nvidia should improve the performance of data and AI workloads on Databricks’ Data Intelligence Platform.
Databricks CEO Ali Ghodsi and Nvidia CEO Jensen Huang announced this during the Data + AI Summit. The two companies have been working together for some time but will now bring CUDA-accelerated computing to the core of the Databricks platform. CUDA is used for accelerating applications, which is essential in generative AI.
“Together with NVIDIA, we’re excited to help every organization build their own AI factories on their own private data,” Ghodsi stated. Added Jensen Huang: “By bringing NVIDIA CUDA acceleration to Databricks’ core computing stack, we’re laying the foundation for customers everywhere to use their data to power enterprise generative AI”
Tip: Databricks enters next phase with major Lakehouse Platform update
Photon
In addition to this collaboration to bring CUDA to the Data Intelligence Platform, Ghodsi also intends to develop native support for Nvidia accelerated computing in the next generation of Databricks vectorized query engine Photon. This should bring more speed and efficiency to data warehousing and analytics workloads. Photon powers Databricks SQL, a serverless data warehouse service. According to Databricks and Nvidia, the collaboration could lead to the next step in price-performance ratio.
Techzine is attending Databricks’ Data + AI Summit this week. Keep an eye on the website for the latest developments on the company.