Moogsoft today launched the Moogsoft Observability Cloud, a software-as-a-service (SaaS) instance of its existing platform for detecting anomalies and prioritizing alerts using AI via machine learning algorithms.
Company CEO Phill Tee said now every IT organization can employ machine learning algorithms to better manage their IT environment without having to set up and manage an artificial intelligence for IT operations (AIOps) platform. Pricing for the Moogsoft Observability Cloud will be based on data ingestion rates for teams and enterprises, but for now, Moogsoft is making available a free trial edition.
A collector ingests time-series metric data directly from sources such as Amazon EC2, Docker, MongoDB, Redis and other platforms. Other metrics can also be ingested via a user-definable metrics API. The goal is to make it possible for IT teams to benefit from AI in minutes rather than having to wait days or weeks for machine learning algorithms to learn the entire IT environment, said Tee.
Designed around a microservices-based architecture and a set of REST application programming interfaces (API), Tee noted the Moogsoft Observability Cloud makes it possible for DevOps teams to self-service their own observability requirements, he added.
The Moogsoft Observability Cloud applies statistical calculations and anomaly detection algorithms to time-series metrics data. A correlation engine then transparently matches patterns to provide more context to a single notification, which Tee noted reduces the overall alert fatigue.
The Moogsoft platform can also be integrated with IT collaboration tools such as PagerDuty and can be configured to send data to any endpoint via an open webhook API. Collectively, that capability enables IT teams to build workflows to get to the root cause of an IT issue faster while simultaneously reducing the noise generated by existing monitoring tools, Tee said.
Moogsoft plans to continue to provide the existing on-premises edition of its platform but expects over time the bulk of IT organizations will prefer to rely on an AIOps platform that is managed on their behalf. Regardless of the delivery model, Tee said IT teams soon will start interacting with these platforms using speech interfaces that will automatically identify pressing issues before they become a major incident.
In the meantime, Tee said resistance to AI is dropping among IT professionals. Not only are more of them encountering various forms of AI used widely in consumer applications, but many IT professionals also have come to realize it will not be possible for a team of humans to optimize complex IT environments without relying more on machine learning algorithms and other forms of AI. In fact, organizations soon may find it difficult to hold on to IT professionals who increasingly expect to have AI tools available to them, he noted.
There’s already no shortage of options when it comes to AIOps platforms. Before too long, every IT management platform will have been infused with machine learning algorithms. The issue now is determining how much those algorithms will forever transform existing IT management processes.