Rightsizing and The Need For A Global Perspective

October 4th, 2013 by

VMTurbo offers a Software Defined Control system that orchestrates virtual environment resources to meet application service levels while utilizing the infrastructure as efficiently as possible.  This orchestration includes a broad range of decisions, including workload placement, capacity (adding or removing), as well as workload rightsizing decisions to maintain the environment in this desired state.

VMTurbo's software controls the virtualized environment in the desired state

A number of products attempt to enable and even automate rightsizing decisions by examining workload configuration and resource usage in isolation.  This approach is flawed, however, and can actually move the environment away from the desired state as defined above.  This is because the workload configuration is impacted not only by its own resources, such as vMem or vCPU, but also by the resources of the underlying hosts and datastores, as well as other workloads, hosts and datastores in the environment.  In other words, the act of resizing a given workload can have “knock on” effects across other virtual machines in the virtual environment, and it is, therefore, imperative to take a global perspective.

The VMTurbo control system analyzes the state of the workload resources based on:

  • Real time and historical performance and utilization of data centers, hosts, virtual machines, datastores, and applications (guest workloads), including peak utilization values
  • Peaks and historical values, which carry a weight to assure workloads are not downsized below peaks
  • The topological dependencies of the virtualized infrastructure (e.g., virtual machines on hosts, or datastores)
  • Configuration and business constraints
  • Workload placement policies
  • The rate of resize in compliance with your business and change management process (low means reach the desired state in conservative increments, where high means reach the desired state in one step)

VMTurbo continuously analyzes all these metrics and topological dependencies across the entire virtualized IT stack and across the entire data center to drive actions (such as workload configuration and placement, resource configuration, allocation and provisioning, and more) to control the environment in the desired state.

Learn more in the Workload Rightsizing white paper.

Leave a Reply

Your email address will not be published. Required fields are marked *