What is Cloud Bursting in Cloud Computing?

Cloud bursting (in cloud computing) is the act of temporarily moving applications into a compute cloud service for processing. The reason it's called "bursting" is because it is short-lived; you move the application for a specified amount of time, and then you move it back to the on-premises servers.

While the concept is fairly simple, the reality of deploying a cloud bursting model is a bit trickier. Applications running in the cloud need to access the same data they would if they were running locally.

To solve this dilemma, you can take several paths. The first option is copying that data into the cloud. However, this involves using different protocols than the native NFS or SMBprotocols that are used in the data center. You may also be "double-paying" if you plan on having the data both in the cloud and on-premises. Alternatively, you can keep the data in the data center and make it available to those applications through a network link. With this approach, however, you will likely run into some serious performance issues. The third option uses Edge filer caching of active data needed for compute into the compute cloud. This helps minimize latency that users experience when running apps and eliminates the need to transfer all of the data onto the cloud.

This 5-minute overview of cloud bursting outlines the basics of how cloud bursting in cloud computing works, explains the challenges in applying this concept, and walks through one method of addressing these problems. In this video, you'll learn:

  • Why it is so economically advantageous to expand compute capacity with the cloud
  • How to maintain performance while running your apps in the cloud
  • Where the vFXT virtual caching layer fits into cloud bursting