Falkonry, Inc. announced the launch of its Falkonry Edge Analyzer while attending the Gartner Data and Analytics Summit. The Edge Analyzer is a portable, self-contained engine that enables customers to deploy predictive analysis on edge devices for low latency applications in disconnected environments or close to data sources. The new Analyzer is available as part of Falkonry’s “pre-packaged” machine-learning system, Falkonry LRS, which enables operations teams to create and deploy predictive analytics in the cloud or on-premises or on the edge without requiring data scientists, according to its maker.
“Many organizations can benefit from a packaged analytics application that includes embedded machine-learning capabilities, especially when pursuing a common and well-defined advanced analytics problem, such as price optimization or fraud detection,” says Peter Krensky, Gartner senior principal / analyst and Carlie Idoine, Gartner senior director / analyst in the March 2018 research piece “How Data Science Teams Leverage Machine Learning and Other Advanced Analytics.”
“Before using Falkonry, companies often found themselves stuck in proof of concepts without a clear path for scaling to production,” said Dr. Nikunj Mehta, founder and CEO of Falkonry. “The ability for operations teams to create and deploy predictive models at the edge and on premises has addressed the integration and skills gap challenges, while realizing five to ten times annual ROI.”
Prediction & explanation
Falkonry LRS’ automated feature-learning solves the most complex problem of applying machine learning on time-series data, according to Falkonry, saving time and building accurate predictive models. The explanation feature gives insight into model results, quantifying signal contribution and enabling SMEs to perform root-cause analysis.
Edge Analyzers can be created in Falkonry LRS and transported for installation in remote or mobile environments. Minimal resource requirements enable operation in constrained environments. Analyzers are configurable for high availability and can tolerate sensor and network outages. Use of containers enables runtime to be insulated from other processing activities. Each Edge Analyzer can be used to monitor multiple edge endpoints, and several Edge Analyzers can be deployed on a single computer to support multiple assessments.
“Real-time asset monitoring and predictive analytics is important to Fluke,” said Oliver Sturrock, CTO of Fluke Digital Systems. “Falkonry’s solution scores high in terms of architecture, scalability and flexibility to deploy in the cloud or at the edge for real-time insights.”