Oracle’s Next-Gen Enterprise AI Services Powered by NVIDIA’s Cutting-Edge GPUs

Oracle and NVIDIA have recently announced an expansion of their partnership to enhance enterprise AI services, making them more accessible, powerful, and practical. This collaboration, unveiled at Oracle AI World, encompasses a wide range of innovations, from groundbreaking new hardware to deeply integrated software designed to embed AI at the heart of a company’s data infrastructure.

Ian Buck, NVIDIA’s VP of Hyperscale and High-Performance Computing, emphasized the significance of this partnership, stating that it paves the way for cutting-edge accelerated computing, streamlining database AI pipelines, accelerating data processing, and facilitating the deployment and scalability of inference on OCI.

The highlight of this collaboration is the introduction of the OCI Zettascale10 computing cluster, powered by NVIDIA GPUs. This platform is optimized for AI training and inference workloads, boasting a remarkable 16 zettaflops of peak AI compute performance. It leverages NVIDIA’s Spectrum-X Ethernet networking fabric to efficiently scale up to millions of processors.

However, the true value of this partnership lies in the software integrations that aim to seamlessly integrate AI into every facet of a business’s operations. The newly introduced Oracle AI Database 26ai embodies this strategy, challenging the conventional approach of moving data to AI models by bringing AI to the data instead. This database release simplifies the integration of AI into data operations, enabling customers to deliver trusted AI insights and innovations across operational systems and analytic data lakes.

Juan Loaiza, Oracle’s Executive VP of Database Technologies, highlighted the simplicity and efficiency of Oracle AI Database, emphasizing its ability to combine private, sensitive data with public information securely within the database environment. The database also implements NIST-approved quantum-resistant algorithms to safeguard data against future quantum computing threats.

The collaboration extends to working with NVIDIA’s toolset, allowing seamless integration with NVIDIA NeMo Retriever for enhanced AI capabilities. Oracle’s Private AI Services Container is set to receive a GPU-powered upgrade, enabling faster data preparation for AI applications.

The partnership also aims to simplify the AI pipeline through the Oracle AI Data Platform, which now includes NVIDIA GPU options and the NVIDIA RAPIDS Accelerator for Apache Spark. These tools are consolidated within the Oracle AI Hub, providing a centralized platform for building, deploying, and managing AI solutions.

Overall, the collaboration between Oracle and NVIDIA is focused on addressing real-world challenges faced by businesses in adopting AI. By integrating hardware, data, and software tools into a cohesive ecosystem, they are paving the way for practical, secure, and scalable enterprise AI solutions.