Early warning systems enhancements

New rate of change algorithms and seasonality support join existing anomaly detection, forecasting and root cause analysis capabilities of cloud-based observability platform.

  • Friday, 17th July 2020 Posted 5 years ago in by Phil Alsop
LogicMonitor has introduced enhancements to its LM Intelligence AIOps early warning system. Its dynamic thresholds functionality, first introduced in December 2019, now includes support for seasonality and rate of change. LogicMonitor’s dynamic thresholds intelligently alert IT teams based on historical performance and newly refined algorithms to help businesses save time, avoid alert fatigue, and surface anomalies sooner to proactively prevent downtime.

 

With the introduction of seasonality support, LogicMonitor’s dynamic thresholds now detect patterns in performance and ensure alerts are triggered for anomalies outside of those patterns. The new rate of change algorithms, added as part of this July 2020 enhancement, detects anomalies in the rate at which a metric value is changing as compared to its normal pattern, to enable users to identify issues before they result in a negative impact to the business.

 

“LogicMonitor’s dynamic thresholds save us time and give us peace of mind, because we know we’re not going to miss something due to a threshold being slightly too high or too low. In addition, dynamic thresholds detect behavioural deviance, which often occurs at a level below static thresholds,” said Wania Konageski, global digital service platform architect at Logicalis. “As a result of these enhancements, we now receive early warnings that strengthen our proactive and preventative capabilities. LogicMonitor’s AIOps capabilities are instrumental to our vision, as an MSP, of providing continuous uptime and a seamless experience to customers.”

 

LogicMonitor’s dynamic thresholds use historical performance to generate an expected range for resources, and alert on anomalies that exceed this expected range. Dynamic thresholds significantly reduce alert noise by silencing notifications for static threshold-generated alerts corresponding to normal performance within the expected range. 

 

“If static thresholds are not set or tuned well, dynamic thresholds will ensure alerts are triggered and silenced appropriately. If static thresholds are tuned well and enabled, dynamic thresholds will still defer to them,” said Tej Redkar, chief product officer at LogicMonitor. “The addition of dynamic thresholds seasonality and rate of change support to the existing anomaly detection, forecasting and root cause analysis features of LM Intelligence mean that LogicMonitor customers now have access to the most advanced AIOps capabilities in the market.”

An examination of how Atlassian’s Rovo and Teamwork Graph introduce AI-driven automation into...
Cequence Security has introduced Agent Personas to provide more granular control over AI agent...
Cybercrime in the financial sector has intensified, with AI posing new challenges. CrowdStrike...
Harness report finds that AI coding tools are being widely adopted in software engineering, with...
Public sector leaders globally are evaluating agentic AI for autonomous task completion as...
Boomi plans to acquire Lunar.dev, aiming to expand capabilities in AI governance and improve...
Feedonomics has introduced ACE, a solution designed to help merchants syndicate product data for...
AHEAD expands its European presence through an acquisition, a senior appointment, and new...