Virtana is expanding Kubernetes cluster support to include infrastructure performance management (IPM) and monitoring capabilities via an artificial intelligence platform for IT operations (AIOps).
Jon Cyr, head of product for Virtana, says the Virtana Platform now provides complete support for containers running on Kubernetes clusters alongside existing support for legacy IT environments.
Via a combination of agentless integrations with Prometheus and the Kubernetes application programming interface (API), the Virtana Platform collects health, utilization and performance metrics from their container environments. The Virtana platform provides visibility into Kubernetes workload performance, maps Kubernetes workloads and relationships, correlates Kubernetes components to underlying physical infrastructure, provides tools for tracking and forecasting Kubernetes performance trends and detects performance anomalies.
At its core, Virtana provides a platform infused with machine learning algorithms to manage hybrid cloud computing environments. Last year the company made its first foray into Kubernetes environments by adding a capability that helps IT teams right-size Kubernetes clusters.
Cyr says the overall goal is to resolve issues faster, de-risk optimization initiatives and reduce cloud costs across both cloud-native and legacy IT environments using a single platform. This approach is more efficient than requiring IT teams to acquire and deploy separate management platforms for each.
Achieving that goal requires an AIOps platform capable of keeping pace with the explosion of Kubernetes clusters that are being deployed alongside legacy systems, notes Cyr. Given the inherent level of complexity, it will not be possible for IT teams to maintain service levels without the aid of an AIOps platform, he adds.
It’s still too early to say how much IT operations can be automated using AI. However, there’s general agreement that AI will play a significant role in automating IT. The more standardized the IT environment is, the easier that becomes to accomplish. The challenge is that cloud-native computing environments tend to be highly dynamic, so it may take more time for an AIOps platform to learn the relationships between all the services that make up these environments.
DevOps teams have been ruthlessly automating processes for years, but with the rise of machine learning algorithms it’s only a matter of time before AIOps is more widely embraced. The issue heading into the new year is determining what impact those platforms will have on DevOps workflows and the overall management of IT in both the short and long term.
Regardless of approach, however, IT teams will need to make a fundamental choice. There is now no shortage of AIOps platforms that could be adopted, but many IT organizations already have management platforms in place that, over time, will be augmented by machine learning algorithms. The decision then becomes whether IT teams should adopt a new AIOps platform or wait for providers of existing IT management frameworks to infuse machine learning algorithms into those platforms.
Acquiring any new IT management platform is always an expensive proposition. In addition to the cost of the platform there are always integration and training costs to be considered. The decision to adopt an AIOps platform is likely to be driven by the urgency of the current IT management challenge.