Dynatrace boosts observability platform with generative AI

Dynatrace has incorporated generative AI into its Davis AI engine to let customers more quickly create dashboards, determine the root cause of incidents, and speed mean time to repair.

Davis AI features causal and predictive AI capabilities now, and with the addition of generative AI, Dynatrace says it will offer customers a third mode of AI that applies natural language capabilities to more easily create dashboards, automate workflows, and complete tasks. Davis CoPilot generative AI will work in collaboration with its causal AI technology, which analyzes real-time data, and its Davis predictive AI models that anticipate future behavior based on past data and observed patterns in the environment.

“We look at billions of dependencies going on in the system. We understand the predictive piece to forecast future behavior, we have root cause analysis that understands how things are connected, and that creates an understanding of the environment and knowing what questions to ask. With generative AI, we can feed this information precisely into an LLM [large language model] and let it generate a recommendation and then route it to the right teams,” says Alois Reitbauer, chief technology strategist at Dynatrace.

“Generative AI makes the product accessible to new audiences, helps them build custom workflows without writing code, and eventually a human being is in the driver seat, it would just take the human way more time to run through all the things going on in the system,” Reitbauer said.

The combination of AI technologies will enable IT teams to more quickly spot trends in behavior and predict performance issues that could impact end-user or customer experiences with applications, Dynatrace asserts.

“The three styles of AI in Davis allow them to work better together,” says Gregg Siegfried, vice president and research analyst at Gartner. Generative AI primarily deals with language, and it “makes it much easier for human operators to deal with these machines if they have a human-like interaction style, which might have been what we lacked before generative AI.”

Dynatrace’s platform monitors a slew of components in an environment to track application performance metrics such as response time and latency. Applying AI and automation to monitor applications makes sense because the amount of data and metrics that need to be collected and analyzed to determine the source of problems exceeds what is humanly possible, Siegfried says. 

The advantage of Davis is that it provides some hints and guidance to IT teams when an application starts to encounter performance problems, Siegfried explains, preventing IT from running down the wrong path to resolve the issue. Now with the popularity of generative AI, Davis will enable IT operators to more easily interact with the AI engine to spot and resolve performance issues.

“Being able to understand the behavior of your software and to be able to put yourself in the shoes of your users is important with the complexity of today’s modern applications,” Siegfried says. “The intelligence is able to parse language and deliver an educated response based on their training in the environment.”

Dynatrace’s Davis platform competes with observability products from New Relic, Riverbed, and Splunk, among others. Many monitoring and observability vendors are exploring how to incorporate generative AI into their software platforms, and this could be how customers first begin to apply generative AI in their environments, according to a recent Gartner IT spending update.

“Generative AI’s best channel to market is through the software, hardware and services that organizations are already using,” said John-David Lovelock, distinguished VP analyst at Gartner, in a statement. “Every year, new features are added to tech products and services as add-ons or upgrades. Most enterprises will incorporate generative AI in a slow and controlled manner through upgrades to tools that are already built into IT budgets.”

Dynatrace offers its Davis AI platform in both on-premises and software-as-a-service (SaaS) models. The expanded Dynatrace Davis AI engine will be available in calendar year 2023 to customers as a core technology within in the Dynatrace platform.