Technology Fridays: Oracle AI Platform Cloud Service Delivers Machine Learning the Oracle Way

Welcome to Technology Fridays! Today, I would like to discuss Oracle’s new entrance in the machine learning space with its AI Platform Cloud Service.

The cloud platform market is becoming the ideal playground for machine learning applications. While platforms such as AWS ML, Azure ML or Google Cloud Ml have dominated the space in its initial iteration, customers has been long expecting Oracle’s to become a relevant player when comes to enable machine learning applications. The reason for the high expectations? Very simple; the enterprise world runs on Oracle databases and perceives an Oracle Cloud-based, machine learning platform as a lower entry point into the data science space. The AI Platform Cloud Service is Oracle’s answer to this very competitive field.

Differently from Oracle’s traditional approach to launch proprietary application development technologies, the AI Platform Cloud Service represents an attempt to bring together the best machine learning stacks in the market using a consistent infrastructure and lifecycle management model. In that sense, the AI Platform Cloud Service allow developers to write programs using different deep learning frameworks such as TensorFlow, Keras or Caffe as well as tools like Jupyter, The support for Jupyter is particularly relevant as it allow developers to implement interactive notebooks that can seamlessly scale using Oracle Cloud’s infrastructure.

To get started with Oracle AI Platform Cloud Service, developers can simply launch instances preconfigured with specific machine learning tools and frameworks. The AI Platform Cloud Service provides support for GPU architectures including novel technologies such as NVidia Pascal model for Tesla GPUs. Also, data science instances come pre-configured with technologies such as cuDNN and CUDA SDK.

To facilitate access to data sources, the integration with different data platforms such as Hadoop or Spark as well as mainstream databases. The platform also integrates with Oracle Cloud data services such s MySQL Cloud Service, Cloud Storage, Big Data Cloud Service, Database Cloud Service among several others. Similarly, the AI Platform Cloud Service leverages the Kafka-based Even Hub to enable low latency, high throughput communication between its components.

Application lifecycle management is one of the core capabilities of the Oracle AI Platform Cloud Service. Data scientists using the platform can train and test models using GPU-accelerated architectures. Once developed, AI models can be deployed to the platform infrastructure on which they can be scaled on-demand or even exported to be used in other runtimes. The development lifecycle can be started from different IDEs or from Oracle Cloud’s command line interface.

Competition?

Oracle AI Platform Cloud Service is entering the ultra-competitive market of machine learning cloud platforms. AWS ML and Azure ML were the first entrants in the space and have certainly captured relevant market share. Google Cloud ML excels at its support for TensorFlow applications while the Alibaba Cloud Machine Learning service enables applications built using different deep learning frameworks. Startups such as BitFusion or Floyd can also be considered competitors of the Oracle AI Platform Cloud Service.

Written by

CEO of IntoTheBlock, Chief Scientist at Invector Labs, Guest lecturer at Columbia University, Angel Investor, Author, Speaker.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store