In a data center, the workloads utilize server hardware resources to varying degrees of demand. It is also necessary to provide consistent workload performance in order to achieve consistent service experience. In most cases, the behaviour of workloads is never known prior to the workload being started. Further, the behaviour will change due to resource contention with other simultaneously executing workloads, making resource management a hard problem.
Dynamic phase detection helps us determine the current phase of the workload, providing a means to identify the resource consumed in each phase and using this information to manage the performance of the workload by suitably allocating resources to workloads that are under performing.
In this session, we will present a workload fingerprinting module implemented within the Watcher framework. We will demonstrate the generation of the fingerprint for a set of workloads running on an OpenStack cluster and visualize the fingerprint.
Registration Opening Soon
Registration will open for you to purchase passes for the OpenStack Summit very soon!
We're excited to see you there!
