atsu operational intelligence
atsu provides unparalleled visibility, historical understanding, and advanced prediction of enterprise NAS (NFS & SMB) I/O activity.
The atsu service starts with simple agents installed on Linux and/or Windows, which collect I/O and process data. Data is sent via Kafka where the atsu service collects, aggregates, and displays job, task, and process-level I/O performance data for the entire compute grid. As jobs and tasks run on the grid, atsu is continually scoring their activities, detecting which tasks are performing poorly or causing grid impact.
After deploying the atsu agents in your HPC environment, you’ll have real-time visibility into all application I/O, regardless of backend storage environment. atsu is the ML-powered solution for scaling challenges in your HPC and NAS environments.
Easily compare I/O usage and characteristics between different Jobs (applications running across multiple hosts).
What will impact your file systems? Understand which jobs are overwhelming which mounts, before it happens.