HP AI workstations equipped with NVIDIA CUDA-X libraries offer a full-stack solution for faster and more efficient generative AI model training.
NVIDIA and HP Inc. have recently announced a collaboration that will see NVIDIA CUDA-X data processing libraries integrated into HP’s AI workstation solutions. This will significantly improve the speed and efficiency of data preparation and processing needed for generative AI development.
Using the NVIDIA CUDA compute platform, CUDA-X libraries are designed to expedite data processing across various data types, such as tables, text, images and video. A key component of this collaboration is the NVIDIA RAPIDS™ cuDF library, which significantly boosts the performance of pandas software—a critical tool for nearly ten million data scientists. By utilizing an NVIDIA RTX™ 6000 Ada Generation GPU instead of a traditional CPU-only system, the cuDF library can increase processing speed by up to 110 times without necessitating any code alterations.
By integrating RAPIDS cuDF and other NVIDIA software into Z by HP AI Studio on HP AI workstations, developers gain access to a complete solution that streamlines data science workflows and facilitates faster model training for generative AI through more efficient data processing.
Transforming data science with GPU acceleration
NVIDIA’s CEO, Jensen Huang, emphasized the importance of enhancing pandas, a key tool for many data scientists working on data processing for generative AI projects. “Accelerating pandas with zero code changes will be a massive step forward. Data scientists can process data in minutes rather than hours, and wrangle orders of magnitude more data to train generative AI models,” he remarked.
HP’s President and CEO, Enrique Lores, also stressed the importance of data science in AI development and the need to provide developers with efficient software and systems. “With the integration of NVIDIA AI software and accelerated GPU compute, HP AI workstations provide a powerful solution for our customers,” he added.
Unified GPU and CPU workflows for seamless development
The NVIDIA RAPIDS cuDF library not only accelerates pandas to run on GPUs but also ensures compatibility with third-party libraries, facilitating a unified approach to both GPU and CPU workflows. This compatibility extends to third-party libraries, simplifying GPU and CPU workflows for data scientists. This allows data scientists to efficiently develop, test and deploy models in production environments.
Additionally, with RTX 6000 Ada Generation GPUs providing 48GB of memory per GPU, HP’s Z8 Fury workstations, which can be equipped with up to four RTX 6000 GPUs, are positioned as some of the most powerful tools available for AI development.
The expected availability of NVIDIA RAPIDS cuDF for accelerated pandas processing on HP AI workstations and PCs equipped with NVIDIA RTX and GeForce RTX GPUs is slated for this month, with further integration into HP AI Studio anticipated later in the year.
Also read:
- Is Your House Equipped to Start a Home Business? What to Consider
- Business Travel Meets Flexibility: Changi Airport T3 Hosts JustCo’s Pay-per-Minute Co-working Solution
- Dell Unveils New Generative AI Solutions for Modern Enterprises
Header Image from Freepik
Press release link: https://nvidianews.nvidia.com/news/nvidia-hp-supercharge-data-science-generative-ai-workstations
