AWS Data Pipeline
Amazon Elastic Map Reduce (EMR) provides a hosted Hadoop framework. The Zementis Universal PMML Plug-in (UPPI) for Hive is certified to run on EMR and, therefore, takes full advantage of the dynamic, scalable, on-demand processing capacity available through the Amazon cloud infrastructure.
Zementis UPPI leverages the Predictive Model Markup Language (PMML) industry standard, not only enabling vendor-agnostic deployment of advanced data mining models but effectively turbo-charging existing models with massively-parallel Big Data processing. Amazon EMR has made numerous improvements to Hive, including direct integration with Amazon S3 which allows UPPI to efficiently access data and predictive models. In addition, UPPI automatically generates the AWS Data Pipeline definition file. Based on this file, it is easy to create an AWS Data Pipeline service that automates the movement of data and launches a desired AWS EMR cluster for scoring data against one or many predictive models.
As AWS Technology Partner, Zementis also provides the ADAPA Scoring Engine in the AWS Marketplace. ADAPA is a single-instance, web-based application with a graphical User Interface that can be launched in minutes and allows any data scientist or analyst to quickly test and deploy PMML-based predictive models for interactive scoring. It is also the choice for real-time transaction processing and business process integration via web services.
To learn more about how you could leverage Zementis products with Amazon’s cloud infrastructure, please contact us.