Marvis, the World's first
Virtual Network Assistant

What if you could chat with your wireless network?

Meet Marvis – the first AI-driven virtual network assistant.  Now you can ask questions and get intuitive answers on par with a wireless expert.

Marvis Benefits

Maximize User Experience

  • Faster troubleshooting means less downtime
  • Proactively fix problems before users know they exist

Prioritize IT Spend

  • Focus IT investments on strategic projects not on troubleshooting
  • Prioritize helpdesk tickets based on scope of impact

Gain New Insights

  • Understand your WLAN without hunting through numerous screens
  • Analyze and correlate date to determine the magnitude and scope of problems

Key Components of Marvis

Simple Natural Language Interface

Marvis uses Natural Language Processing (NLP) and Information Theory to analyze massive amounts of data and draw subtle inferences. Simply ask Marvis a question and it will help you extract insights from the system or troubleshoot an issue for you.

Anomaly Detection

Marvis adds anomaly detection to the Mist SLE dashboard so that administrators can rapidly and proactively identify service impacting events that assure rapid determination and resolution of the root cause of issues. Anomaly detection leverages data science tools to automatically determine service baselines and trigger notifications when there is a service-impacting deviation from that baseline. With our API driven interface, detected anomalies can even trigger external events such as creation of a help desk ticket, without manual intervention.

Accurate Root Cause Analysis

Bayesian Inference, a part of our data science toolbox, is used to identify causes with the highest probability of association to the problem occurring on the network. This delivers more accurate root cause analysis to speed up problem identification and resolution.

Correlate Data to Understand Scope

Marvis correlates information across a large knowledge base to determine the scope and magnitude of a problem. This helps you prioritize issues and assign resources efficiently.

For example:

  • Did an upgrade cause a wireless issue?
  • Is it specific to client, group of clients, entire site, or many locations?
  • Is it happening rarely, occasionally, or frequently?
  • Is there a coverage issue or is a DHCP server responding slowly?
  • What is the scope and impact of the problem?

Integrated Help Desk

Marvis uses machine learning to perform unique troubleshooting and helpdesk functions like anomaly detection, event correlation, and confidence ratings. By leveraging APIs, this can trigger automated workflows so you can  rapidly solve (or avoid) wired, wireless and device problems.

Additional Resources