The job scheduler is the central point for your HPC infrastructure. It dispatches and monitors the jobs on the available resources, keeping track of the jobs' allocation and their state. The generated data are stored in the job scheduler's accounting logs, from which valuable information can be extracted in order to understand the cluster behavior.
For example, the analysis of the logs makes it possible to identify practices and historical events that influence the overall cluster performance: which users are requesting more resources; which kind of jobs wait more time in the queue before starting; resources consumption per job state, and so on. Understanding the historical behavior of the cluster is crucial to choose the proper actions towards increasing production and profitability.
That is why we have developed Analyze-IT. By processing the job-submission historical data in the logs, Analyze-IT creates detailed analyses of the behavior of the jobs on your cluster. Moreover, it delivers a report containing indicators and insights about your HPC infrastructure.

This 'AWS Budget' version of Analyze-IT features a costs estimation plugin that allows you to estimate the budget you will need to run a selected workload on AWS, including cost for: Services, Compute, Storage, Remote Visualization and Data-Transfer.