People often think of the Environmental Protection Agency as a primarily regulatory organization, but EPA also provides critical response capabilities in the event of a disaster with an environmental impact. One of those capabilities is measuring the severity of that impact, which can put an added burden on systems with the increased data collection and analysis. That led Vaughn Noga, EPA’s chief information officer, to look for new ways to meet that situational need.

“Several years ago, with the wildfires in California, we had some performance issues associated with one of the air applications,” Noga said on Federal Monthly Insights — Mastering IT Resilience: Strategies for Federal Continuity. “And so we made a concerted effort to really focus on that and to look at how can we increase performance and availability. And that certainly led us right to host that in the cloud environment. It allows us to scale. So obviously, from a seasonal perspective, we don’t need the horsepower or the compute power all the time. So we can scale up and scale down based on demand. And so we actually, within a year of having some of the issues that we saw when it was hosted on prem, we turned that around within one year. And we’re able to actually meet the performance requirements that the public really demanded.”

EPA currently has a “large percentage” of applications in the cloud, Noga said, including mission-oriented applications like this example, as well as many back-office administrative applications as well, including its contracting and human resources systems. EPA has adopted commercial software-as-a-service offerings for some of these, like Relativity, an eDiscovery program EPA uses for its Freedom of Information Act functions.

Noga said EPA currently splits its applications between cloud hosting, and on premises at its National Computer Center in Raleigh, North Carolina.

Proactive monitoring and response

That National Computer Center also hosts part of the integrated team — along with EPA’s Washington, D.C. headquarters — that manages the network operations, security and hosting functions for EPA’s systems. Noga stressed that that integrated nature is important to monitoring for, identifying and resolving issues.

“So it’s not just a network person looking for whether it’s a network issue affecting the application performance or the availability,” Noga told The Federal Drive with Tom Temin. “It’s the entire team that’s looking at the event, if you will, to understand, is it a networking issue? Can we resolve the networking issue or is it a system issue that needs to go to the sysadmins to resolve? And certainly they use both open source and other monitoring tools to make sure that they are aware of what’s going on.”

That’s especially important, Noga said, because EPA is now looking at logging and monitoring from an enterprise perspective. That means data is coming in from multiple servers, networks, systems and even regions.

“It’s a huge data problem. And so the goal really is to find the needle in the stack of needles. And so a lot of this is developing the right rules, looking for the right types of issues so our analysts can really focus on that one event that they need to be focused on,” Noga said. “You really require those types of rule sets be implemented, so you can actually see when something goes bump in the night and so you can actually take action on it.”

Noga said that includes plenty of automation, ranging in complexity from simple alerts triggered by specific events to artificial intelligence tools that can not only spot anomalies in the full catalog of log files, but also correlate them across multiple interfaces.

“We couldn’t do what we do and provide the level of monitoring without the support of those types of tools,” he said.

Copyright © 2024 Federal News Network. All rights reserved. This website is not intended for users located within the European Economic Area.