Overview:
Litware, Inc. is a company that manufactures personal devices to track physical activity and other health-related
data.
Litware has a health tracking application that sends health-related data from a user’s personal device to
Microsoft Azure.
Litware has three development and commercial offices. The offices are located in the United States,
Luxembourg, and India.
Litware products are sold worldwide. Litware has commercial representatives in more than 80 countries.
Existing Environment:
In addition to using desktop computers in all of the offices, Litware recently started using Microsoft Azure
resources and services for both development and operations.
Litware has an Azure Machine Learning solution.
Litware recently extended its platform to provide third-party companies with the ability to upload data from
devices to Azure. The data can be aggregated across multiple devices to provide users with a comprehensive
view of their global health activity.
While the upload from each device is small, potentially more than 100 million devices will upload data daily by
using an Azure event hub.
Each health activity has a small amount of data, such as activity type, start date/time, and end date/time. Eachactivity is limited to a total of 3 KB and includes a customer identification key.
In addition to the Litware health tracking application, the users’ activities can be reported to Azure by using an
open API.
The developers at Litware perform Machine Learning experiments to recommend an appropriate health activity
based on the past three activities of a user.
The Litware developers train a model to recommend the best activity for a user based on the hour of the day.
Requirements:
Litware plans to extend the existing dashboard features so that health activities can be compared between the
users based on age, gender, and geographic region.
Minimize the costs associated with transferring data from the event hub to Azure Storage.
Litware identifies the following technical requirements:
Data from the devices must be stored for three years in a format that enables the fast processing of date
fields and filtering.
The third-party companies must be able to use the Litware Machine Learning models to generate
recommendations to their users by using a third-party application.
Any changes to the health tracking application must ensure that the Litware developers can run the
experiments without interrupting or degrading the performance of the production environment.
Activity tracking data must be available to all of the Litware developers for experimentation. The developers
must be prevented from accessing the private information of the users.
When the Litware health tracking application asks users how they feel, their responses must be reported to
Azure.
Users report that when they access data that is more than one year old from a dashboard, the response time is
slow.
You need to resolve the issue that causes the slow response when visualizing older data.
What should you do?
A.
Process the event hub data first, and then process the older data on demand.
B.
Process the older data on demand first, and then process the event hub data.
C.
Aggregate the older data by time, and then save the aggregated data to reference data streams.
D.
Store all of the data from the event hub in a single partition.