As datasets sizes are growing, a single computer may not be capable to perform data mining algorithms, on the dataset at once. It may happen because of the limited memory, or because it would take too long because of the processing power limitation. There are various methods to overcome this limitation.

The goal of this assignment is to understand and use cloud/distributed databases. Luckily, from a user perspective, those database management systems often look very similar to the local database you may have worked with in other courses.

Directions

1. Literature supported explanation on when and why to use cloud databases and when to use ones managed on a single-device one.

2. Description (in your own words) of how a cloud database management system distributes data between multiple physical devices.

3. Register* with one of the cloud services (e.g. Google Cloud, Microsoft Azure, MongoDB) and create a cloud-based relational or NoSQL database. Upload any dataset related to your project topic into this database. Run at least two sample queries (e.g. aggregation, filtering, e.t.c) to query this database.

Provide evidence of completion of this task (e.g. screenshots).

Note that some services provide free trial accounts for students (e.g. Azure for students(Links to an external site.), or Google Cloud for Students,(Links to an external site.)MongoDB student pack(Links to an external site.)). Unlike some regular trials, some of those accounts do not allow to spend more than a free credit which prevents you from being billed. Saying that please be cautious what credit have you received, what is the billing schema. And delete the resources you created in the cloud hosting after completing your assignment, then you won't spend more credits or being billed in the future.

View Rubric