Intel and AWS, a subsidiary of Amazon that provides on-demand cloud computing services, have worked together for over a decade to ensure AWS services run on a platform optimized for customer workloads at the best value. At the annual AWS re:Invent conference from November 26-30, AWS is highlighting advancements in this ongoing collaboration, including the new AWS Marketplace for Machine Learning, developed by AWS for its Machine Learning (ML) as a Service (MLaaS) platform, Amazon SageMaker*.
Intel offers several AI software tools and libraries on AWS Marketplace, including some of the most popular algorithmic optimizations from the Intel® Data Analytics Acceleration Library (Intel® DAAL) and from BigDL, a distributed deep learning library for Apache Spark*.
The AWS Marketplace for Machine Learning will let sellers create algorithms and models in Amazon SageMaker and list them on AWS Marketplace. For end users, AWS Marketplace makes it easier than ever to access Intel AI software optimized for AWS.
Globally, the volume of data is growing exponentially. Genomics, risk, social network, and consumer preference analysis are just a few examples where high-performance analysis of large datasets is critical in today’s compute landscape. To process increasingly large data sets faster, AWS and Intel are focusing on increasing algorithm performance and accessibility to machine learning applications.
The powerful data analysis solutions of Intel DAAL algorithms such as the decision forestclassifier offer performance optimizations for data ingestion and algorithmic compute together. Additionally, BigDL is an open source distributed deep learning framework for Apache Spark that makes deep learning more accessible by allowing developers to use familiar tools and infrastructure to build deep learning applications. Using BigDL, developers can write deep learning applications as Scala* or Python* programs while leveraging the power of scalable Spark* clusters. This new compatibility with SageMaker adds seamless research-to-production capabilities while retaining the ease-of-use that has enabled deep learning libraries to rapidly gain popularity.
For most of these tasks, computational speed is a key ingredient for success. Intel AI software offerings on the new AWS Marketplace for Machine Learning help developers reduce the time it takes to develop high-performance applications and deploy them at scale. Intel DAAL enables applications to make better predictions faster and analyze larger data sets with available compute resources. This new Amazon SageMaker deployment option promises to combine the performance boost of Intel DAAL and other software optimizations with the easy deployment of Amazon SageMaker so that customers can build, train, and deploy machine learning models quickly and easily.
With the pre-configured environment within Amazon SageMaker, developers and data scientists can specify their scripts using a single API call to train locally or to submit a distributed training job. With a second API call, developers can also now deploy Intel DAAL-trained models to a managed, highly available, online endpoint that can be automatically scaled up or down as demand requires. Developers can easily change parameters and take advantage of Amazon SageMaker features like automatic model tuning. This allows developers to:
When the model is validated after training and tuning, it can be deployed to a managed, highly available, online endpoint in Amazon SageMaker, with automatic scaling, integrated monitoring and logging, and high throughput capabilities.
Intel’s collaborative hardware and software development efforts have powered AWS innovations in cloud infrastructure for years. These newly-announced AI software integrations with Amazon SageMaker will allow developers and data scientists to leverage the infrastructure and simplicity of AWS services with the performance acceleration of Intel-optimized algorithms.
Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration. No computer system can be absolutely secure. Check with your system manufacturer or retailer or learn more at intel.com.
Intel and the Intel logo are trademarks of Intel Corporation or its subsidiaries in the U.S. and/or other countries.
*Other names and brands may be claimed as the property of others.
© Intel Corporation