Skip to main content

Multiple Datastores Working Together: Will It Blend?

Alex Brandt (Rackspace Hosting)
Databases & Datastores
Portland 256
Average rating: ***..
(3.17, 6 ratings)
Slides:   1-PDF 

There has been an explosion in datastore technologies. Whereas developers had to choose from a few technologies in the past, there exists a wide array of datastore options today. There are five main types of datastores: Relational, Column Family, Graph, Key-Value and Document. Polyglot Persistence, or the ability to have many different types of datastores interacting with one application, is becoming more prominent and beginning to take center stage.

A multitude of datastores has led to each kind specializing in data storage and retrieval. Consequently, developers are able to get more out of the datastore than ever before; by relying on the strengths of a particular datastore they can reduce the amount of code needed to accomplish a task. Consequently, developers are able to get more out of the datastore than ever before, reducing the amount of code needed to accomplish a task. However, it is now more important for developers to take a critical look at which datastore to house their applications’ data.

Furthermore, applications are not necessarily restricted to one datastore. Having multiple datastores allows developers to take advantage of the strengths inherent and unique to each storage technology.

In my presentation I plan to cover the following topics:

  • A call for a renewal of critical analysis when choosing a datastore
  • Overview of the different types of datastores, the strengths of each and a discussion on when to consider the different options
  • Walk through use cases for combining different datastore technologies

At the end of the presentation, participants will see that choosing a datastore for their application is not a paint-by-the-numbers proposition, but requires significant thought and knowledge of the myriad options available today. Attendees will understand that while some frameworks lean heavily on particular datastores, it is important to think critically about the needs of the application and the data to ensure that the correct datastore is chosen for the job.

Photo of Alex Brandt

Alex Brandt

Rackspace Hosting

Alex Brandt is a cloud developer at Rackspace, helping people build distributed scalable systems with a variety of technologies. An emerging evangelist for all things cloud, he has worked in IT support and solutioning as well as research in the realm of physics. Alex holds a B.S. in computer science and physics from Minnesota State University Moorhead.