Cisco preps technology to predict enterprise-network problems

Cisco says it is working on a service to let enterprises proactively avert network problems and increase performance.

The company says it has built a predictive analytics engine it will offer via software-as-a-service (SaaS) to help network operators quickly and accurately predict network issues and prevent problems before they happen.

“The future of connectivity will rely on self-healing networks that can learn, predict and plan,” Chuck Robbins, Cisco chair and CEO said in a statement. “Our research for predictive networks has been tested and developed with customers, and early adopters [including Phillips 66, Schneider Electric and the Adecco Group] are seeing major benefits saving them time and money.”

Unplanned downtime and outages can interrupt employee productivity, customer service, and revenue, and as the networks expand to keep up with demands, so does the threat landscape, Cisco stated. 

“Most businesses may think they are surviving this to an extent, but the truth is enterprises need a holistic network watchdog that can detect hardware issues, bandwidth spikes and app config changes before they happen, with high accuracy,” Cisco stated.

According to Cisco, its AI/ML-based predictive engine works by gathering telemetry data across an organization from routers, switches, servers, and more. Once integrated, the engine aggregates the data, learns patterns using a variety of models and predicts user experience issues and provide concrete remediation options. 

The engine looks at connectivity and quality of service when predicting and planning alternative routes. Customers can decide how widely they want to connect the engine in the network, giving them flexible expansion options, Cisco stated.  

The company said it has been building and testing predictive software engines over the past two years and ultimately will offer it across its range of products. 

More detail is expected at its upcoming Cisco Live! customer event in June.