NVIDIA GPUs Fuel Oracle’s Next-Gen Enterprise AI Services

A key element of this partnership is the introduction of the Oracle Cloud Infrastructure (OCI) Zettascale10 computing cluster. This platform is accelerated by NVIDIA GPUs and is tailored for demanding AI training and inference workloads.

The OCI Zettascale10 boasts a peak AI compute performance of 16 zettaflops. Critical to its scalability is NVIDIA’s Spectrum-X Ethernet, a networking fabric engineered to minimize GPU idle time, allowing organizations to efficiently scale AI workloads across millions of processors. This addresses a primary bottleneck in large-scale AI deployments: data throughput and inter-GPU communication.

Beyond raw compute power, the partnership emphasizes software integrations to integrate AI into various facets of business operations. This strategic move could potentially disrupt traditional AI development workflows.

Mahesh Thiagarajan, Executive VP of Oracle Cloud Infrastructure, highlighted the capabilities of the new platform. “OCI Zettascale10 delivers multi‑gigawatt capacity for the most challenging AI workloads with NVIDIA’s next-generation GPU platform. In addition, the native availability of NVIDIA AI Enterprise on OCI gives our joint customers a leading AI toolset close at hand to OCI’s 200+ cloud services, supporting a long tail of customer innovation.”

Giving Your Oracle Database a Brain with AI

The cornerstone of this strategy is the Oracle AI Database 26ai. Oracle is challenging the conventional approach of data-to-model movement, by advocating bringing AI to the data, emphasizing enhanced security and efficiency. This new database release is a tangible manifestation of this “AI for Data” philosophy, potentially reshaping how enterprises approach AI integration with their data assets.

Juan Loaiza, Executive VP of Oracle Database Technologies at Oracle, stated, “By architecting AI and data together, Oracle AI Database makes ‘AI for Data’ simple to learn and simple to use. We enable our customers to easily deliver trusted AI insights, innovations, and productivity for all their data, everywhere, including both operational systems and analytic data lakes.”

A key feature is the ability to implement agentic AI workflows directly within the database. These AI agents are designed to tackle sophisticated queries by leveraging a blend of enterprise’s private data with public information, all without data leaving the secure environment. Key enablers include a Unified Hybrid Vector Search, for context discovery across varying data (relational tables, JSON files, spatial maps).

Oracle is also addressing long term security concerns. The new database incorporates NIST-approved quantum-resistant algorithms designed to protect data in transit and at rest. This proactive approach defends against “harvest now, decrypt later” attacks, where encrypted data is stolen now, hoping to decrypt it with future quantum computing breakthroughs. This positions Oracle favorably amidst growing concerns regarding the quantum threat to data security.

The new database integrates with NVIDIA’s software suite, allowing programming interfaces to seamlessly connect to NVIDIA NeMo Retriever, a collection of microservices aimed to streamline complex AI workflows.

This integration simplifies the implementation of retrieval-augmented generation (RAG). RAG allows language models to access and process information within company documents before generating answers, leading to more relevance and accuracy with responses.

The Oracle Private AI Services Container is also receiving a GPU-powered upgrade. The container allows companies to operate AI models securely within existing infrastructure. Vector embeddings, a fundamental component of AI search, can be offloaded to NVIDIA GPUs using the cuVS library, significantly reducing data preparation time for AI applications.

Democratizing Enterprise AI

Beyond the database, the partnership seeks to streamline the entire AI workflow. The Oracle AI Data Platform now includes an integrated NVIDIA GPU option and the NVIDIA RAPIDS Accelerator for Apache Spark. This directly benefits data scientists and engineers, empowering them to leverage GPUs for faster data processing and machine learning workflows, often requiring minimal code modifications.

All of these components are being consolidated within the Oracle AI Hub, designed to serve as a single interface for building, deploying, and managing AI solutions. Within the hub, users can implement NVIDIA’s NIM microservices through a simplified interface.

The full NVIDIA AI Enterprise software suite is now natively available within the OCI Console. Developers can initiate a GPU instance and enable NVIDIA tools with ease, eliminating procurement bottlenecks. This streamlined onboarding boosts team efficiency when starting new AI projects.

This collaboration addresses the prevalent challenges in enterprise AI adoption. By coalescing hardware, data infrastructure, and software tools, Oracle and NVIDIA are promoting a future where practical, secure, and scalable enterprise AI becomes the norm.