Juniper Networks Expands Mist AI to the WAN

Juniper Networks on Wednesday said its AI-driven enterprise now extends to the WAN. This is the fourth expansion of the Juniper Mist technology that brings artificial intelligence to the LAN, WLAN, security and now, the WAN.

Juniper Mist WAN Assurance is part Juniper’s AI-driven enterprise strategy, or the client to cloud automation experience. This experience begins with WI-Fi Assurance. This makes Wi-Fi predictable, reliable and measurable.

Next is Wired Assurance. It delivers reliable wired experiences. In February, Juniper announced updates to its Juniper Connected Security strategy. Referred to as a threat-aware network, Juniper offers deep network visibility and security policy enforcement to all connection points across the network.

Christian Gilby is senior director of product marketing with Mist, a Juniper company.

Mist Systems’ Christian Gilby

“This fourth iteration of the AI-driven enterprise is about the WAN,” Gilby told Channel Futures. “So we’re extending Mist AI to the WAN, or SD-WAN space, which is another challenging area as you look at networking.”

That challenge is businesses having connections for multiple branch sites going over the internet, and getting insight to understand the user experience across that link.

“From the network evolution perspective, we started from the edge and now we’re working our way in,” he said.

WAN Assurance Shifts Focus to the User

Juniper Mist WAN Assurance brings AI-driven insights for the WAN and branch to Juniper’s secure SD-WAN solution. The goal is to shift the focus from network and application behavior to actual user experiences.

Here’s more about the Juniper Mist WAN Assurance cloud-based service:

• It streams key telemetry data from Juniper SRX devices to the Mist AI engine in the cloud to enable customizable WAN service levels that are used to optimize user experiences.
• It works in conjunction with Marvis, the virtual network assistant, to correlate events across the LAN, WLAN and WAN. This provides rapid fault isolation and resolution across all domains.
• It enables proactive actions as a result of anomaly detection with automated workflows, with an eye toward completely self-driving networks from client to cloud.

The extension to Marvis is a conversational interface that streamlines operations. The Marvis enhancement uses two new technologies — natural language understanding and knowledge graphs. The knowledge graph is a data science tool that allows Marvis to navigate the data sets faster.

“Our view on the conversational interface is that ultimately this can evolve to become part of an IT team,” said Gilby. “If you look at trends in the market today, there’s mobile — that’s been a trend for quite some time; IoT; and there are more devices coming into the enterprise. And it’s more challenging to support those [devices]. IT teams are pressured on budget. And you can’t double the number of support staff as you double the number of devices coming in. That’s where we believe AI is critical.”

More About Marvis

Courtesy: Juniper Networks

Here are the enhancements to Marvis, the interactive IT assistant for insight and troubleshooting.

• Deduces the intent behind the questions and provides valuable results based on its access to a large knowledge base. This includes interactive queries to drill down further based on suggested next steps.
• Allows users to give feedback on whether Marvis answered the right question or gave the correct root cause. This allows Marvis to continuously learn, which is a key requirement for AI-driven support.
• Built using open APIs, allowing it to communicate with existing chatbots as needed.

Dartmouth College uses Marvis. As a result, the school saves 90% in …