Building Media & Entertainment Predictive
Analytics Solutions on AWS
December 2016
© 2016, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Notices
This document is provided for informational purposes only. It represents AWS’s
current product offerings and practices as of the date of issue of this document,
which are subject to change without notice. Customers are responsible for
making their own independent assessment of the information in this document
and any use of AWS’s products or services, each of which is provided “as is”
without warranty of any kind, whether express or implied. This document does
not create any warranties, representations, contractual commitments,
conditions or assurances from AWS, its affiliates, suppliers or licensors. The
responsibilities and liabilities of AWS to its customers are controlled by AWS
agreements, and this document is not part of, nor does it modify, any agreement
between AWS and its customers.
Contents
Introduction 1!
Overview of AWS Enabled M&E Workloads 1!
Overview of the Predictive Analytics Process Flow 3!
Common M&E Predictive Analytics Use Cases 6!
Predictive Analytics Architecture on AWS 8!
Data Sources and Data Ingestion 9!
Data Store 13!
Processing by Data Scientists 14!
Prediction Processing and Serving 22!
AWS Services and Benefits 23!
Amazon S3 23!
Amazon Kinesis 24!
Amazon EMR 24!
Amazon Machine Learning (Amazon ML) 25!
AWS Data Pipeline 25!
Amazon Elastic Compute Cloud (Amazon EC2) 26!
Amazon CloudSearch 26!
AWS Lambda 26!
Amazon Relational Database Service (Amazon RDS) 26!
Amazon DynamoDB 26!
Conclusion 27!
Contributors 27!
Abstract
This whitepaper is intended for data scientists, data architects, and data
engineers who want to design and build Media and Entertainment (M&E)
predictive analytics solutions on AWS. Specifically, this paper provides an
introduction to common cloud-enabled M&E workloads, and describes how a
predictive analytics workload fits into the overall M&E workflows in the cloud.
The paper provides an overview of the main phases for the predictive analytics
business process, as well as an overview of common M&E predictive analytics
use cases. Then, the paper describes the technical reference architecture and
tool options for implementing predictive analytics solutions on AWS.
Amazon Web Services – Building Media & Entertainment Predictive Analytics Solutions
Page 1
Introduction
The world of Media and Entertainment (M&E) has shifted from treating
customers as mass audiences to forming connections with individuals. This
progression was enabled by unlocking insights from data generated through new
distribution platforms and web and social networks. M&E companies are
aggressively moving from a traditional, mass-broadcasting business model to an
Over-The-Top (OTT) model where relevant data can be gathered. In this new
model, they are embracing the challenge of acquiring, enriching, and retaining
customers through big data and predictive analytics solutions.
As cloud technology adoption becomes mainstream, M&E companies are moving
many analytics workloads to AWS to achieve agility, scale, lower cost, rapid
innovation, and operational efficiency. As these companies start their journey to
the cloud, they have questions about common M&E use cases and how to design,
build, and operate these solutions. AWS provides many services in the data and
analytics space that are well suited for all M&E analytics workloads, including
traditional BI reporting, real-time analytics, and predictive analytics.
In this paper, we discuss the approach to architecture and tools. We’ll cover
design, build, and operate aspects of predictive analytics in subsequent papers.
Overview of AWS Enabled M&E Workloads
M&E content producers have traditionally relied heavily on systems located on-
premises for production and post-production workloads. Content producers are
increasingly looking into the AWS Cloud to run workloads. This is due to the
huge increase in the volume of content from new business models, such as on-
demand and other online delivery, as well as new content formats such as 4k and
high dynamic range (HDR).
M&E customers deliver live, linear, on-demand, and OTT content with the AWS
Cloud. AWS services also enable media partners to build solutions across M&E
lines of business. Examples include:
• Managing digital assets
• Publishing digital content
• Automating media supply chains