Oracle OpenWorld, San Francisco, Calif—Oct 2, 2017
Expanding its comprehensive Artificial Intelligence (AI) offerings, Oracle today announced Oracle AI Platform Cloud Service to help developers quickly create and deploy breakthrough enterprise AI services. Using the new cloud service, organizations can take advantage of deep learning to better understand enterprise data and transform corporate business processes and user experiences.
Only Oracle can deliver a pervasive approach to AI, embedding AI into its applications, providing an AI development environment coupled with a world-class infrastructure to run specialized workloads.
The complexity and time required to set up a complete environment for work on advanced machine learning algorithms can be daunting. With today’s announcement, Oracle is making it quick and easy for developers and data scientists to set up a secure, scalable environment for new deep learning models in the cloud. Oracle Cloud has been uniquely engineered to streamline model training by leveraging customers’ essential enterprise data and operating on an industry leading high-speed network, object storage, and GPUs. Oracle AI Platform Cloud instances come pre-installed with familiar AI libraries, tools, and deep learning frameworks, including Caffe, Jupyter Notebook, Keras, NymPy, scikit-learn, and TensorFlow, among others. Machine learning practitioners also have access to Oracle Object Store and can easily connect to existing Spark/Hadoop clusters.
“AI has the power to be more transformative for the enterprise than any other technology in recent history,” said Amit Zavery, senior vice president, product development, Oracle Cloud Platform. “Oracle is in a unique position to deliver AI across all layers of the cloud, empowering customers to uncover and unlock critical business patterns in their enterprise data to transform organizational productivity, efficiency, and insight.”
Oracle AI Platform Cloud Service runs on world-class Oracle Cloud Infrastructure optimized for peak performance and price-performance to run specialized deep learning applications. The differentiated infrastructure layer includes NVMe flash storage and best-in-class GPUs on a 25 Gigabit network for extreme performance. In an industry standard benchmark test, Oracle was 2X more performant than Azure’s largest GPU instance and demonstrated 2.4X better price-performance than the closest AWS equivalent.(1) Oracle AI Platform Cloud Service includes a GPU Bare-Metal Shape: 2x Tesla P100 GPUs based on Nvidia’s Pascal Architecture, and will soon support the new Volta GPUs with up to 8 GPUs.
Once machine-learning models are trained, developers can access Oracle’s rich set of PaaS services to include them in AI-powered applications.
Adaptive Intelligent Apps are powered by insights from the Oracle Data Cloud, which is the largest third-party data marketplace in the world with a collection of more than 5 billion global consumer and business IDs and more than 7.5 trillion data points collected monthly. By applying advanced data science and machine learning to Oracle’s web-scale data and an organization’s own data, the new Adaptive Intelligent Apps can react, learn and adapt in real time based on historical and dynamic customer data.
Oracle’s holistic approach enables customers to quickly employ advanced machine learning to transform their business processes. In addition to Oracle AI Platform Cloud Service, Oracle embeds ready-to-use AI and machine learning capabilities across Oracle’s SaaS, PaaS, and IOT services, including cognitive AI, analytics, data services, IT management, and security operations. Adding machine learning and cognitive interactions to traditional business processes and applications enables greatly improved user experience and productivity.
Other new AI-based solutions and services being unveiled or showcased at Oracle OpenWorld include:
The Oracle Cloud offers complete SaaS application suites for ERP, HCM and CX, plus best-in-class database Platform as a Service (PaaS) and Infrastructure as a Service (IaaS) from data centers throughout the Americas, Europe and Asia. For more information about Oracle (NYSE:ORCL), please visit us at oracle.com.
Oracle OpenWorld, the industry's most important business and technology conference for the past 20 years, hosts tens of thousands of in-person attendees as well as millions online. Dedicated to helping businesses leverage Cloud for their innovation and growth, the conference delivers deep insight into industry trends and breakthroughs driven by technology. Designed for attendees who want to connect, learn, explore and be inspired, Oracle OpenWorld offers more than 2,500 educational sessions led by more than 2,000 customers and partners sharing their experiences, first hand. With hundreds of demos and hands-on labs, plus exhibitions from more than 400 partners and customers from around the world, Oracle OpenWorld has become a showcase for leading cloud technologies, from Cloud Applications to Cloud Platform and Infrastructure. For more information; to register; or to watch Oracle OpenWorld keynotes, sessions, and more, visit www.oracle.com/openworld. Join the Oracle OpenWorld discussion on Twitter.
Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners.
The preceding is intended to outline our general product direction. It is intended for information purposes only, and may not be incorporated into any contract. It is not a commitment to deliver any material, code, or functionality, and should not be relied upon in making purchasing decisions. The development, release, and timing of any features or functionality described for Oracle's products remains at the sole discretion of Oracle Corporation.
Statements in this article relating to Oracle’s future plans, expectations, beliefs, and intentions are “forward-looking statements” and are subject to material risks and uncertainties. Such statements are based on Oracle’s current expectations and assumptions, some of which are beyond Oracle’s control. All information in this article is current as of October 2, 2017 and Oracle undertakes no duty to update any statement in light of new information or future events.
 Based on industry-standard SPECint_rate2006 benchmarking, a measure of multicore, multisocket workloads. Compared to AWS M4.16xlarge
 Based on a standard benchmark with TensorFlow as the main Neural Net Training Framework, AlexNet and Inception V3 computer vision models, and the ImageNet dataset. Compared to AWS p2.8xlarge and Azure NC24