Oracle’s Next-Generation AI and ML services and NVIDIA are pushing the boundaries of innovation

March 22, 2022 | 3 minute read
Ashish Sharma
Director Partner Product Marketing
Text Size 100%:

Oracle’s Next-Generation AI and ML Services Push Boundaries of Innovation at NVIDIA GTC

Oracle Cloud Infrastructure (OCI) is proud to participate in NVIDIA GTC 2022, where we’re showcasing examples of how Oracle AI combined with the power of NVIDIA accelerated computing enables a new breed of AI-powered cloud and high-performance computing (HPC) applications and pushes the boundaries of innovation across industries and use cases.

Data is the new fuel and HPC is the engine

Data is often referred to as the new oil, not only because of its potential value to businesses, but also because data is critical in training AI and machine learning models. In the AI field, ML algorithms are only as good as the quality and quantity of data they’re trained on. So, the computational power of NVIDIA GPUs are crucial for processing large volumes of data and delivering on the promise of AI.

Oracle delivers key ingredients to jumpstart successful AI projects that are ready to use—pretrained ML services that can be seamlessly integrated by developers, data scientists, and even business analysts directly.

Oracle AI and ML services for developers and data scientists

OCI AI services are a collection of prebuilt machine learning models that developers can easily add to their applications and business operations. These models, which you can further custom train on an organization’s own business data, are pretrained on industry data, which helps them deliver more accurate results. Developers can now focus on accelerating application development without needing data science experience.

The following services all run directly on NVIDIA GPU offerings in OCI:

Unified AI/ML platform spanning cloud services, apps, and data assets

For data scientists who want to build custom machine learning models or solve business problems that OCI AI services don’t cover, Oracle provides developer platforms that data scientists can use to build end-to-end machine learning models and put to work. These platforms include OCI Data Science and Machine Learning in Oracle Database, both of which support  NVIDIA GPUs to accelerate every step of the ML lifecycle from model building to deployment.

To learn more about our collaboration with NVIDIA, watch Ian Buck, GM and Vice President of Accelerated Computing at NVIDIA, and Greg Pavlik, Senior Vice President and CTO of Oracle Cloud Platform, discuss how to make AI work for you.

Powerful and accelerated for all

AI is no longer limited to the realm of large enterprise projects. Through low pricing models for OCI AI and ML services with NVIDIA GPUs, innovation in AI is emerging from non-profit organizations, academia, and startups, including the following examples:

  • Children´s Medical Research Institute CMRI cut costs by 25% while enabling AI-based medical research and gene therapy using OCI Data Science, OCI Vision, and NVIDIA GPUs.
  • University of Oxford researchers built a predictive deep-learning model by analyzing over 1 million job ads using OCI Data Science and NVIDIA GPUs.
  • DeepZen creates audiobooks in a record time using their AI-based text-to-speech solutions running on Oracle Cloud and NVIDIA GPUs.

Supporting the latest NVIDIA Hopper architecture on Oracle Cloud Infrastructure

Today, NVIDIA announced the NVIDIA H100 Tensor Core GPUs, based on the new NVIDIA Hopper architecture, enable an order-of-magnitude leap for large-scale AI and HPC workloads. Oracle looks forward to supporting the latest NVIDIA technology. The world’s largest and most powerful accelerator, the NVIDIA H100, is ideal for training large AI language models, deep recommender systems, genomics, building complex digital twins, and other HPC applications.

Learn more about how our joint customers are innovating with AI and HPC at our sessions at GTC. For more information and to try an Oracle AI workshop, visit Artificial Intelligence (AI) from Oracle. To read more about Oracle Cloud Infrastructure (OCI)’s investment in building the most innovative and best price-performant cloud has enabled our customers to take on greater challenges that require advanced AI and computing capabilities, check out Andrew Butterfield's Oracle Cloud is racing ahead at NVIDIA GTC 2022 blog post.

Ashish Sharma

Director Partner Product Marketing

Previous Post

Announcing the general availability of Oracle Cloud Infrastructure Operations Insights for bare metal, virtual machines, and Exadata Cloud service

Sriram Vrinda | 3 min read

Next Post

Establishing connectivity from an Oracle Cloud VMware Solution NSX-T overlay segment to another VCN in different region

Vaibhav Tiwari | 4 min read