Optimization: The Fourth Key to Success in Complex Compute Environments

Welcome back to the “Seven Keys to Success” blog series. Last time we talked about compute environments’ need for interaction by users and administrators through manual steps that are often repetitive and time-consuming. We discussed ways in which organizations can save time, efficiently utilize resources, and speed individual and end-to-end processes through the automation of those manual steps. In this post we will introduce the fourth key to success – Optimization.

Resource Timeline

Unfortunately, nearly all compute environments operate below their maximum potential. Moab helps organizations get the most out of their system by optimizing run time, power usage, license usage, and cost reduction, enabling them to get more work done at a lower cost:

  • Run Time – Moab’s time-aware, predictive scheduling takes information about resources, current and reserved workloads, planned maintenance, deadlines, service commitments, and wait time, in order to model out into the future how to best pack workloads. Additional capabilities like dynamic prioritization and intelligent preemptions ensure that run time is optimized for the most important tasks.
  • Power – With its power management policies, Moab is able to automatically lower the power state of idle nodes in order to reclaim excess power consumption, all while maintaining system responsiveness. It can also automate the optimal CPU clock frequency on a per-application basis, thereby minimizing energy costs and helping organizations stay within power caps.
  • License – Through license and workload awareness, including knowledge of wait times, Moab provides the information necessary for organizations to determine the optimal quantity of licenses for their needs. It can also intelligently spread out workload over time to get the most out of licenses already owned. Moab makes this process easy as it integrates with FlexLM to provide a dynamic license management experience.
  • Cost – With Moab, resource costs can be set for different hardware, storage, and other factors, with its scheduling algorithm achieving the lowest cost that will still meet SLA objectives. Show-back and charge-back policies help organizations to naturally align usage behavior and reduce costs per users, groups, and projects.

Moab helps compute environments achieve their full potential by dynamically optimizing workload processing, thereby reducing costs and improving the speed of producing end results.

Series Navigation<< Automation: The Third Key to Success in Complex Compute Environments