Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?

 
ON THIS PAGE
 

Overview of Juniper Mist Wi-Fi Assurance

Juniper Mist™ Wi-Fi Assurance is a cloud service based on machine learning (ML) and driven by Mist Al™. It replaces manual troubleshooting tasks with automated wireless operations to make Wi-Fi predictable, reliable, and measurable. Juniper Mist Wi-Fi Assurance provides network administrators full visibility into the user experience on their wireless networks.

Mist, AI is in the air. How good is your Wi-Fi network right now? Can you prove it? Traditional wireless LANs focus on network and AP uptime, but up doesn't mean good, and a lack of complaints doesn't mean a lack of problems. Your users are showing you exactly what they need for a great wireless experience.

For the first time, there's a platform that actually listens to and learns from them. Mist started with a clean sheet of paper. We looked at how big data companies correlate massive amounts of information using the cloud and machine learning, and asked, how do we harness these same principles to build the first user-centric wireless network? We designed our own control plane that exposes over 100 user states in real time for unprecedented visibility.

Each transition generates valuable metadata that represents the entire user experience. Mist's powerful purpose-built cloud architecture collects and processes this information with unparalleled performance and scale. These user insights have actually been there all along.

Mist is the first to liberate this data from wireless chipsets and legacy architectures. With Mist, Wi-Fi is finally predictable, reliable, and measurable. Set, monitor, and enforce service levels based on what's right for your users, not on arbitrary vendor metrics.

Track thresholds to ensure service levels are being met in real time and trending in a positive direction. If any users experience a problem, Mist will show the root cause in just one click. Issues are correlated across the wired, wireless, and device domains for easy identification and possible remediation.

Drill down to see if issues are occurring during specific periods of time or are a result of recent system changes. With Mist, radio resource management is no longer a black hole because you see the exact impact that changes have on the user experience. Mist also lets you spot trends across device types, operating systems, and applications, pinpointing which areas need immediate attention and which are interesting but not critical.

Identify troublesome APs and or WLANs so you can spend your time where it matters most. All of these insights are possible because you now have real visibility into the user experience. Visualize user states over time and inspect problematic events.

If a service-impacting event occurs, Mist triggers a dynamic PCAP and stores it in the cloud for easy analysis. No more wasting time and money chasing temporal problems with expensive truck rolls and sniffers. Predictive recommendations help avoid issues from happening in the future, replacing reactive troubleshooting with proactive wireless operations.

All of Mist's capabilities are 100% programmable via open APIs. Remotely configure and activate a new site in just minutes. Automate the overnight testing of point-of-sale devices.

Create alerts notifying when baseline metrics aren't being met or are negatively trending. There's no limit to the types of services that are now possible using Mist automation. Mist assures that your network is performing at its peak and your users are happy.

With Mist, the Wi-Fi experience is finally predictable, reliable, and measurable. This is the new wireless network designed for the smart device era. Mist, AI is in the air.

Juniper Mist™ Wi-Fi Assurance helps you keep the focus on your users' network experience, with built-in machine-learning (ML) features that tune the network for the best possible user experience. Network administrators can set service thresholds for time to connect, wireless coverage, throughput, and more. When user experience doesn't match these service-level expectations (SLEs), Juniper Mist tells you why. When problems arise, our AI-driven virtual network assistant, Marvis, shows you the root cause and often suggests action to fix the problem.

Juniper Mist Wi-Fi Assurance includes access points (APs), Mist Edges, wireless LANs (WLANs), radio management, security, network segmentation, and a lot of automation.

Scalable Cloud-Based Architecture

The Juniper Mist platform is built on a modern microservices cloud architecture, which offers the scalability to meet your network needs. The architecture supports multitenancy and scales with the elasticity of the cloud. Thus, a managed service provider (MSP) can operate wireless sites and networks for dozens of client organizations. A large retailer can accommodate various site-specific requirements yet manage the organization centrally, with a single login per cloud. Other features and benefits include:

  • Microservices—The Juniper Mist cloud is built on a microservices architecture that brings agility and scale to network management and operations. On-demand network upgrades and patches take minutes instead of months. Services are designed independently of one another. These features lay the foundation for a flexible, scalable, resilient, programmable, and agile network. See What Is a Cloud Microservice for more information.

  • Programmability through open APIs—Juniper Mist is 100% programmable with all functions available through open APIs. Every task you can perform in the Mist GUI you can perform through an API call. For a complete, searchable list of APIs that you can try out for yourself, see the Mist API Reference.

  • Controllerless management—Wi-Fi assurance runs in the cloud. You don't have any on-premises controllers to install or manage, so your network scales easily.

  • Centralized data path—When an organization needs to retain a centralized data path architecture for security in campus or branch deployments, the Juniper Mist Edge solution provides that data path. The use of the centralized data path enables Juniper Mist to operate like a controller-based solution, in that the on-site data paths are secure. The tunneling functions of Juniper Mist Edges let Juniper Mist keep control and management functions securely in the Juniper Mist cloud.

  • Performance optimization and regulatory compliance—To comply with data residency regulations and optimize performance, we deploy Juniper Mist cloud instances worldwide. Server locations include Europe, Asia, and Australia. In the United States of America (USA), we have servers on the east and west coasts as well as a USA federal cloud.

MIST. AI is in the air. MIST is the first vendor in over a decade to bring true innovation to the wireless space.

By leveraging a modern cloud architecture with microservices, the MIST Learning WLAN eliminates the challenges of legacy wireless controllers and the shortcomings of early cloud solutions. First introduced in 2003, controllers are hardware devices that run a single monolithic software image. Over time, more and more features were added to the controller's common code base, creating interdependencies that greatly increased the complexity of the software.

Development and testing cycles became longer, engineering teams got larger, and software bugs grew out of control. This all changes if your platform is built on the modern cloud. Complex applications are now split into core functions called microservices, each of which are designed by a small and focused development team.

Services are designed independently of one another using optimized technology stacks chosen specifically for that service. Adding or removing features is simple, and bugs are fixed in near real-time, without network disruption. Services scale up or down elastically when they're needed, without requiring expensive hardware.

And the platform is inherently resilient, as the failure of one service does not impact the others. And since this is all software, we can help customers realize cloud-like agility at the edge. Most business applications have already moved to the more efficient modern cloud.

Now wireless can too, leaving outdated controllers and their archaic software for the history books. Agility. Flexibility.

Scalability. Intelligence. This is the new wireless network built on the modern cloud.

MIST. AI is in the air.

Centralized Configuration and Operations

The Juniper Mist portal is entirely API driven and provides configuration options for every aspect of your wireless network. You can perform a wide range of tasks on the Juniper Mist portal:

  • Create users

  • Define and apply policies

  • Define and apply WLAN templates to your sites

  • Define and apply device profiles that can override template settings

  • Set up sites within your organization

  • Investigate and solve problems with the help of AI

With APIs behind every function, you can programmatically control every aspect of your network. If you are GUI-minded, you can use the Juniper Mist portal. For the automation experts and application integrators, there's the Mist API Reference.

Templates and Device Profiles

In the Juniper Mist portal, you'll often find that you can make the same configuration settings in different places. For example, you can configure radio resource management (RRM) and other radio settings directly on the Juniper access points (APs), in a device profile, or an RF template. The modular design makes it easy to scale configurations across different AP groupings or quickly assemble any combination of APs, WLANs, access policies, and radio frequency (RF) configurations.

Note:

We recommend that you use templates, device profiles, and other group configuration options to simplify the management of your devices and networks in the Juniper Mist portal. Using these configuration aids prevents direct device or network configuration, which can be error-prone and difficult to track.

For details, see: Configure a WLAN Template and RRM Configuration Options.

WxLAN Policies

The Wireless extensible LAN (WxLAN) policies in Juniper Mist allow you to define organization and site-level labels. You can apply these labels to many types of objects, including users, WLANs, APs, IP addresses, IP subnets, applications, and so on. You can easily apply WxLAN policies to endpoints, regardless of their authentication or authorization method. With labeling and visual policy design, you replace multiple lines of CLI syntax with easy-to-read graphic policy displays.

As shown in the image above, WxLAN policies allow for granular configuration of the rules within the policy. Juniper Mist processes the rules from the top down. Juniper Mist applies the first rule that matches the traffic and does not process any more rules. See WxLAN Access Policies for more details.

Flexible and Secure Guest Wi-Fi Access

Guest access is an important consideration for any enterprise WLAN. Juniper Mist provides options to customize your WLAN for guest access, including:

  • Scalable guest access solution—Your guest access solution can incorporate flexible options including multiple language support, customizable branding, social login, external captive portal integration, and RADIUS integration. See WLAN Guest Portal.
  • WLAN segmentation—Segment your guest WLAN by using multiple preshared keys (PSKs) and making each PSK a personal WLAN. There's no connectivity between devices with different PSKs.
  • Guest traffic—You can locally bridge guest traffic, tie it to a dedicated guest Ethernet port, or tunnel it to a centralized concentrator from your APs.

See Compare WLAN Guest Portal Options for more details.

Proactive Optimization of Wireless Performance

Juniper Mist-managed WLANs undergo periodic optimization as part of our radio resource management (RRM) implementation. Juniper Mist uses a once-daily RRM optimization based on machine learning (ML) data and applies this optimization to all access points (APs) in a site. You can also trigger individual frequency band optimization at any time, or when you make changes to certain settings.

Juniper Mist offers the following services to improve wireless performance:

  • Customizable wireless service levels—You can set and monitor service-level expectation (SLE) metrics for key performance areas. Juniper Mist gathers these metrics from all client devices. The SLEs determine what constitutes success in your network, including connection times, throughput, and more. The Wireless Capacity SLE plays a key role in the RRM optimizations mentioned above. See Service-Level Expectations (SLE).
  • AI-driven RRM—Juniper Mist applies data science and reinforcement learning strategies to the aggregate SLE performance data to:
    • Optimize radio settings.
    • Assure network performance.
    • Adapt to radio interference.
    • Respond to environmental issues.

    Juniper Mist applies the previous actions, as appropriate, to all sites within your organization.

    For more information, see Radio Management.

  • Marvis Virtual Network Assistant (VNA)—The Marvis VNA proactively identifies issues, interprets the scope and magnitude of the impact, identifies the root causes, and recommends fixes. In some cases, Marvis can take action to correct issues without your intervention.

  • Marvis Minis—Marvis Minis are network digital twins that validate network and application services for your network even when no clients are connected to the network. Marvis Minis detect issues and assess the impact of those issues. So, they'll tell you if the problem impacts a switch, an AP, a WLAN, or a server.

  • Digital transformation with network insights—The Juniper Mist Wi-Fi Assurance cloud service includes base capabilities for analyzing up to 7 days of data, simplifying the process of extracting network insights from data across your enterprise.

Wireless Service-Level Expectations

Juniper Mist service-level expectations (SLEs) help you understand a user's wireless network experience. Here's the typical SLE workflow:

  1. Juniper access points (APs) collect key data about every user's wireless experience and upload it to the Juniper Mist cloud.

  2. The Juniper Mist portal normalizes the data to a user-minute metric.

  3. Juniper Mist applies machine learning (ML) to create measurable, actionable information about successful connections, time to connect, wireless throughput, and more.

This process helps determine whether the user experience was good or bad during the last user minute. And, if it was bad, it identifies the SLE that was not met."

At any time, you can see how your network is performing against the wireless SLEs with deep visibility into impacted users, applications, and devices. From the various Juniper Mist dashboards, you can visualize this information across the entire organization, individual sites, or even individual clients.

From the dashboard, you can visualize the data for the entire organization, individual sites, or even individual clients.

As you can see in the preceding image, the filter is currently set to show Wireless SLE metrics for today on the Live Demo site. You can also configure the scope of the visualization for a single site, an entire organization, and specific APs or clients as shown below.

For more information, see Wireless SLEs.

Troubleshooting

When questions or problems arise, Juniper Mist provides you with the following automated tools that help answer your questions or solve your problems.

  • Dynamic packet capture—The Juniper Mist Wi-Fi Assurance cloud service automatically starts capturing packets when it detects specific anomalies. Juniper Mist captures more than 150 state changes for each client device and access point every few seconds. With this record, you can rewind in time to see what, exactly, was going on when the event occurred. This eliminates hours of guesswork or time spent trying to reproduce an issue. See Dynamic and Manual Packet Captures for more information.
  • Marvis Actions—Use the predictive recommendations and automated workflows in Marvis Actions to quickly solve problems or prevent them from occurring.
  • Root-cause identification—Using data science and machine learning, the Proactive Analytics and Correlation Engine (PACE) aids in the root-cause identification of problems so you can quickly identify and fix them.