Massively Scaling Computing Power

There is always a reason to enlarge your private OpenStack cloud environment—successful deployment. Preparing a layout design that's ready to grow on demand is quite challenging. For this reason, OpenStack has been designed to grow seamlessly. Procurement of additional resources as needed should be straightforward. That is where capacity planning management best practices come into play. It is essential to ensure that any request to the OpenStack resource pools should be served without limitation. On the other hand, from an infrastructure perspective, the available hardware that runs an OpenStack private cloud will always have capacity limits including computing, networking, and storage resources. Taking into account the latter consideration, we will look at the fundamental building blocks of OpenStack (compute service) in detail and push the limits of computing power in your private cloud. In this chapter, we will cover the following points:

• Discussing the compute service in detail and adding a new compute node using Ansible
• Listing and configuring supported hypervisors in OpenStack, including Docker and Xen
• Defining new approaches on how to scale the compute service by leveraging a few OpenStack terminologies regarding compute cluster segregation
• Learning the mechanism of scheduling and weighing in OpenStack to boost the compute workload