Experience Monitoring with Path Analysis (EA)

This article explains how accounts with a DEM license can use Path Analysis to better understand digital experience for their sites and users.

Note

Note: This is an Early Availability (EA) feature that is only available for limited release. For more information, contact your Cato Networks representative or send an email to ea@catonetworks.com.

Overview

Experience Monitoring lets you track the digital experience of your users as they interact with their applications and services. You receive an overall Application Experience score for your experience, which can help you identify problems or trends in your network. For more information, see What is Cato Experience Monitoring.

DEM-overall.png

In addition, for accounts with a DEM license, Cato provides tools that give you additional insight into the user and site experience. Path analysis helps you break down the network path into logical nodes and highlight possible problem indicators.

With Path Analysis, you have a visualized breakdown of the different points of the flows for a site and user. Depending on the type of site and the location of the user you see different nodes. For example, for a user working remotely, you can view information about your device, Wi-Fi, LAN gateway, Internet, tunnel, and application experience.

path-analysis-user.png

Whereas for a host without the Cato Client you will see information only for the Internet, tunnel and application.

path-analysis-socket.png

For information about the different Experience Monitoring widgets, see Using the Experience Monitoring Page.

Use Case

A user from ABC company calls the IT helpdesk complaining that the company's messaging system is responding poorly.

When looking at the overall Application Score for the messaging system, the helpdesk engineer doesn't see any global issue with the application. However, when looking at the User Path Analysis, they see that the Device node is colored yellow.

path-analysis-usecase.png

Hovering over the Device node, they see that while the CPU usage is around 27%, Memory usage is over 65%. This helps the helpdesk engineer understand that the problem probably lies in a local problem on the device and not a problem in the network or messaging application.

EA Known Limitations

  • Scores are based on successful TCP flows. Blocked traffic is not taken into account.

  • Traffic sent using the Bypass feature does not provide HTTP/s Latency and Error Rate metrics

  • Data in the Experience Monitoring widgets is displayed with an approximately 30-minute delay

Node Scoring

Each node is scored for its individual experience based on metrics for that node. When looking at the Path Analysis, you can immediately get an indication of the score for each node based on its color (green, yellow, and red). Each node is scored individually and doesn't affect the score of other nodes.

In addition, when you hover over a node, you can view the average metrics for that node. To see the full metrics, click on the node to view the relevant widgets.

The following table provides information about which nodes are available, the metrics used to calculate the score, and how the score is calculated. For information about Application Experience Scoring, see Application Experience Scoring.

Note

Note: Device, Wi-Fi, and LAN Gateway data is supported for Windows Client v5.11 and macOS Client v.5.7

Depending on the connection type, you will see different nodes on the Path Analysis.

Node

Metrics

How it's scored

Connection Type

Client Device

The Client Device score is based on the following metrics, as shown in the widgets:

  • CPU Score

    • Good - CPU in use is less than 70%

    • Fair - CPU in use is between 70-90%

    • Poor - CPU in use is greater than 90%

  • Memory Score

    • Good - Memory used is below 60%

    • Fair - Memory used is between 60-80%

    • Poor - Memory used is greater than 80%

The score displayed for the node is based on the least good result between the two metrics.

For example, if the CPU Score is Good, but the Memory Score is Fair, the node score is Fair.

  • Remote User

  • User behind Socket site using Cato Client

  • User behind an IPsec site using Cato Client

  • User behind Cloud Interconnect site using Cato Client

Wi-Fi

The Wi-Fi node score is calculated based on the Signal Strength metric, as shown in the widgets.

  • Good - signal strength above -67 dBm.

  • Fair - signal strength above -80 dBm

  • Poor - signal strength below -80 dBm

The score displayed for the node is based on the score for the Signal Strength metric

  • Remote User

  • User behind Socket site using Cato Client

  • User behind an IPsec site using Cato Client

  • User behind Cloud Interconnect site using Cato Client

LAN Gateway

The score for the LAN Gateway is based on the following metrics, as shown in the widgets:

  • LAN Gateway Packet Loss

    • Good - less than 1% packet loss.

    • Fair - between 1-5% packet loss.

    • Poor - more than 5% packet loss.

  • LAN Gateway Distance

    • Good - below 10 ms.

    • Fair - between 10-50 ms.

    • Poor - exceeding 50 ms

The score displayed for the node is based on the least good result between the two metrics.

For example, if LAN Gateway Distance is Good and LAN Gateway Packet Loss is Poor, the node score is Poor.

  • Remote User

  • User behind Socket site using Cato Client

  • User behind an IPsec site using Cato Client

  • User behind Cloud Interconnect site using Cato Client

Internet

The score for the Internet node is based on the following metrics, as shown in the widgets:

  • Internet Packet Loss

    • Good - less than 1% packet loss.

    • Fair - between 1-5% packet loss.

    • Poor - more than 5% packet loss.

  • Internet Distnace

    • Good - below 20 ms.

    • Fair - between 20-100 ms.

    • Poor - exceeding 100 ms

The score displayed for the node is based on the least good result between the two metrics.

For example, if Internet Distance is Fair and Internet Packet Loss is Good, the node score is Fair.

  • Remote User

  • Socket site

  • Host behind a Socket site not using Cato Client

  • User behind Socket site using Cato Client

Tunnel

The score for the Tunnel node is based on the following metrics, as shown in the widgets:

  • Tunnel Upstream Packet Loss

    • Good - less than 1% packet loss.

    • Fair - between 1-5% packet loss.

    • Poor - more than 5% packet loss.

  • Tunnel Downstream Packet Loss

    • Good - less than 1% packet loss.

    • Fair - between 1-5% packet loss.

    • Poor - more than 5% packet loss.

  • Tunnel Distance

    • Good - below 20 ms.

    • Fair - between 20-100 ms.

    • Poor - exceeding 100 ms

In addition, there is also a Tunnel Age widget that is not calculated in the score.

The score displayed for the node is based on the best result between the two metrics.

For example, if the Tunnel Packet Loss is Good, and Tunnel Distance is Poor, the node score is Good.

  • Remote User

  • Socket site

  • IPsec site

  • Host behind Socket site not using Cato Client

  • Host behind an IPsec site not using Cato Client

  • User behind Socket site using Cato Client

  • User behind IPsec site using Cato Client

Application

The application score is based on the current application score and any applied filters.

All connection types

Was this article helpful?

1 out of 1 found this helpful

0 comments

Add your comment