NVIDIA and HP Inc. have joined forces to integrate NVIDIA CUDA-X™ data processing libraries with HP AI workstations, enhancing data preparation and processing for generative AI development.

These CUDA-X libraries, built on the NVIDIA CUDA® compute platform, significantly accelerate data processing across various types, such as tables, text, images, and video. The inclusion of the NVIDIA RAPIDS™ cuDF library, for instance, boosts the performance of data scientists using pandas software by up to 110x with an NVIDIA RTX™ 6000 Ada Generation GPU, without necessitating any code modifications.

With RAPIDS cuDF and other NVIDIA software available through Z by HP AI Studio, HP AI workstations offer a comprehensive solution for expediting data science workflows.

Enrique Lores, HP Inc.’s president and CEO, emphasized the importance of fast access to software and systems for AI development, while Jensen Huang, NVIDIA’s founder and CEO, highlighted the significant advancement enabled by accelerating pandas without code changes.

This integration is particularly beneficial for data scientists working with large datasets, as the RTX 6000 Ada Generation GPUs provide ample memory and processing power to handle complex AI workloads effectively. The collaboration between HP and NVIDIA aims to empower data scientists to streamline their development processes and tackle even the most demanding generative AI tasks efficiently.

4x faster nvidia rtx tensorrt llm 2

The accelerated pandas functionality through RAPIDS cuDF is expected to be available on HP AI workstation solutions equipped with NVIDIA RTX and GeForce RTX GPUs starting this month, with broader availability on HP AI Studio later this year.

Leave a Reply

Your email address will not be published