When you read the current literature on cloud systems management, one key factor is what is called “observability.”  With so many moving parts, one tends to focus on a specific group of indicators and miss the overall activity.

When Vivek Kundra started to talk about “Cloud First” back in 2009. He had no idea the size and complexity of clouds that would evolve in his desire to reduce cost and increase flexibility for federal projects.

SolarWinds has been a leader in system observability for decades. Today, we have a person who has successfully used Solar Winds on a variety of systems and relates some of the best practices for gaining this elusive observability.

Scott Pross has seen systems managers look at specific aspects of federal systems in silos. For example, a manager may have data coming in on servers, or even the network itself. Another set of metrics may give information on applications. Managing virtual environments has evolved into a category in and of itself.

When a problem arises, a troubleshooter may microfocus on an area, but not realize how it impacts the entire ecosystem. Scott suggests that using solutions from SolarWinds can give a systems analyst a view from 40,000 feet instead of ten feet off the ground.

Another consideration is how to monitor activity in a cloud when its architecture is changing. The change may be for compliance reasons, expanding applications, or something as basic as running out of room and having to adapt to an influx of data.

Leaders of federal agencies are not interested in extremely detailed observations about logs.  They want to know when their application will be available for citizens or employees. Addressing this issue, Scott details how, using SolarWinds, he can assemble dashboards to give leaders a better understanding of what has gone wrong to enable them to make data-based decisions.