Welcome to Virtualization and Beyond

Contributed by Michael Thompson, Director, Systems Management Product Marketing, SolarWinds

App-Centric and Admin-Centric—Too Much to Ask?

With all the technology trends thrown around this time of year, there is a pleasant consistency that applications or application-centric capabilities that meet real end user needs are critical. Whether it's the cloud, hyper-convergence or mobility, it's all about providing an application service in a better way.

This application-centric focus is especially evident in the IT management space, where there are a lot of vendors thriving because they are providing insight into application performance in a better way than their predecessors did. For example, Splunk has taken log management into new territory, New Relic has taken byte code instrumentation to new environments and use cases, and we at SolarWinds provide industrial strength infrastructure and application monitoring solutions with an entirely new cost and usability level versus prior solutions.

However, most vendors still haven't been successful at providing an application management solution that is truly admin-centric. A solution that is both app-centric and admin-centric would be one that brings together multiple perspectives of an application problem that mirrors the way an administrator would operate when trying to solve the problem. For example, an administrator might find a problem while using a transaction monitoring tool or when an application monitoring alert is triggered. Their next step might be to go to their application monitoring tool to get initial information. Next, they might look at all the infrastructure under that application. If through this process they can narrow down the problem to a specific area or component, they will then need to pull correlated logs potentially related to the event.

That's just one example out of a pretty long list of technical approaches that each provide a different perspective on application performance. In addition to application and infrastructure monitoring, log and event correlation and byte code instrumentation, there is also real and synthetic transaction monitoring. These technologies provide the foundation for rapid root cause analysis and improved mean time to repair, but to go the next step it needs to include the correlation between the technologies. As an added bonus, wouldn't it be nice to have something like a collaboration tool that allow work to be shared between people, too?

While the tools available today to do each of these tasks have gotten substantially easier to use, there is still much room for improvement in bringing all the information together from the various sources.

Today, many vendors have brought together two or three different application management or troubleshooting tools in context of the application, but given the pressure on IT to improve performance with less, the race is on to pull multiple data streams together to get a holistic view of application performance for optimization and troubleshooting.

About the Author

Michael Thompson, Director, Systems Management Product Marketing, SolarWinds.