Automation: The Third Key to Success in Complex Compute Environments
Welcome back to the “Seven Keys to Success” blog series. Previously, we considered the effects of siloed environments and disjointed disciplines on the performance of complex compute environments, and discussed the roles of convergence and consolidation in meeting those challenges. In this post we will introduce the third key to success – Automation.
Functions of compute environments regularly require manual interaction by users and administrators in both the pre and post steps of the creation of individual workloads in addition to the manual steps between workloads which are part of a larger end-to-end process. This dependency on human involvement, in tasks which are often repetitive and time-consuming, can end up wasting large portions of time, resources, and money. Moab lessens or eliminates this need for manual involvement by users and administrators through the application of powerful job dependency and trigger capabilities.
Job dependencies are able to automate the launching of workloads in conjunction with other workloads based on success or failure. This is necessary, for example, when the results of one task are needed by a subsequent task for a refined analysis. Automation by job dependencies effectively turns the links between these workloads into accelerated end-to-end processes, thereby increasing efficiency and reducing unnecessary costs.
Moab’s trigger capabilities allow it to build out workflow steps, which may even include elements that are not directly part of the workload. Triggers can automate actions based on combinations of data, object states, events, metrics, counters, service levels, or environmental conditions. They can be attached to workloads, nodes, storage, networks, or reservations. For example, with triggers Moab can turn on nodes, run health checks, stage data, or perform security scrubs on nodes after a workload is complete.
In compute environments requiring wasted manual steps by users and administrators, Moab’s automation is key to speeding individual and end-to-end processes and saving resources.