###BeginCaseStudy###
Case Study: 4
Lucerne Publishing
Background
Overview
Lucerne Publishing creates, stores, and delivers online media for advertising companies. This
media is streamed to computers by using the web, and to mobile devices around the world by
using native applications. The company currently supports the iOS, Android, and Windows
Phone 8.1 platform.
Lucerne Publishing uses proprietary software to manage its media workflow. This software
has reached the end of its lifecycle. The company plans to move its media workflows to the
cloud. Lucerne Publishing provides access to its customers, who are third-party companies,
so that they can download, upload, search, and index media that is stored on Lucerne
Publishing servers.
Apps and Applications
Lucerne Publishing develops the applications that customers use to deliver media. The
company currently provides the following media delivery applications:
• Lucerne Media W – a web application that delivers media by using any browser
• Lucerne Media M – a mobile app that delivers media by using Windows Phone 8.1
• Lucerne Media A – a mobile app that delivers media by using an iOS device
• Lucerne Media N – a mobile app that delivers media by using an Android device
• Lucerne Media D – a desktop client application that customer’s install on their local
computer
Business Requirements
Lucerne Publishing’s customers and their consumers have the following requirements:
• Access to media must be time-constricted once media is delivered to a consumer.
• The time required to download media to mobile devices must be minimized.

• Customers must have 24-hour access to media downloads regardless of their location
or time zone.
• Lucerne Publishing must be able to monitor the performance and usage of its
customer-facing app.
Lucerne Publishing wants to make its asset catalog searchable without requiring a database
redesign.
• Customers must be able to access all data by using a web application. They must also
be able to access data by using a mobile app that is provided by Lucerne Publishing.
• Customers must be able to search for media assets by key words and media type.
• Lucerne Publishing wants to move the asset catalog database to the cloud without
formatting the source data.
Other Requirements
Development
Code and current development documents must be backed up at all times. All solutions must
be automatically built and deployed to Azure when code is checked in to source control.
Network Optimization
Lucerne Publishing has a .NET web application that runs on Azure. The web application
analyzes storage and the distribution of its media assets. It needs to monitor the utilization of
the web application. Ultimately, Lucerne Publishing hopes to cut its costs by reducing data
replication without sacrificing its quality of service to its customers. The solution has the
following requirements:
• Optimize the storage location and amount of duplication of media.
• Vary several parameters including the number of data nodes and the distance from
node to customers.
• Minimize network bandwidth.
• Lucerne Publishing wants be notified of exceptions in the web application.
Technical Requirements
Data Mining
Lucerne Publishing constantly mines its data to identify customer patterns. The company
plans to replace the existing on-premises cluster with a cloud-based solution. Lucerne
Publishing has the following requirements:
Virtual machines:
• The data mining solution must support the use of hundreds to thousands of processing
cores.
• Minimize the number of virtual machines by using more powerful virtual machines.
Each virtual machine must always have eight or more processor cores available.
• Allow the number of processor cores dedicated to an analysis to grow and shrink
automatically based on the demand of the analysis.
• Virtual machines must use remote memory direct access to improve performance.
Task scheduling:
The solution must automatically schedule jobs. The scheduler must distribute the jobs based
on the demand and available resources.
Data analysis results:
The solution must provide a web service that allows applications to access the results of
analyses.
Other Requirements

Feature Support
• Ad copy data must be searchable in full text.
• Ad copy data must indexed to optimize search speed.
• Media metadata must be stored in Azure Table storage.
• Media files must be stored in Azure BLOB storage.
• The customer-facing website must have access to all ad copy and media.
• The customer-facing website must automatically scale and replicate to locations
around the world.
• Media and data must be replicated around the world to decrease the latency of data
transfers.
• Media uploads must have fast data transfer rates (low latency) without the need to
upload the data offline.
Security
• Customer access must be managed by using Active Directory.
• Media files must be encrypted by using the PlayReady encryption method.
• Customers must be able to upload media quickly and securely over a private
connection with no opportunity for internet snooping.
###EndCaseStudy###

You need to recommend an appropriate solution for the data mining requirements.
Which solution should you recommend?

A.
Design a schedule process that allocates tasks to multiple virtual machines, and use the Azure
Portal to create new VMs as needed.

B.
Use Azure HPC Scheduler Tools to schedule jobs and automate scaling of virtual machines.

C.
Use Traffic Manager to allocate tasks to multiple virtual machines, and use the Azure Portal to
spin up new virtual machines as needed.

D.
Use Windows Server HPC Pack on-premises to schedule jobs and automate scaling of virtual
machines in Azure.

Explanation:
* Microsoft Azure Traffic Manager allows you to control the distribution of user traffic to your
specified endpoints, which can include Azure cloud services, websites, and other endpoints. Traffic
Manager works by applying an intelligent policy engine to Domain Name System (DNS) queries for
the domain names of your Internet resources. Your Azure cloud services or websites can be running
in different datacenters across the world.
* Scenario:
Virtual machines:
The data mining solution must support the use of hundreds to thousands of processing cores.
Minimize the number of virtual machines by using more powerful virtual machines. Each virtual
machine must always have eight or more processor cores available.
Allow the number of processor cores dedicated to an analysis to grow and shrink automatically
based on the demand of the analysis.
Virtual machines must use remote memory direct access to improve performance.
Task scheduling:
The solution must automatically schedule jobs. The scheduler must distribute the jobs based on the
demand and available resources.

https://azure.microsoft.com/sv-se/documentation/articles/traffic-manager-overview/