ExtremeAnalytics Network Service Dashboard


To access the Network Service Dashboard, open the Analytics > Dashboard tab and select Network Service in the dashboard drop-down list.

Network Service Dashboard

Overview

The Network Service Dashboard contains two graphs for each network service: the Expected Response Time bar graph displays the average response time over the selected time period and the Historical Response Time line graph displays the individual response times over that period for each site.

Select the number of sites displayed in each column in the Top field.

Use the Time Period drop-down list to display the date and time range for which data is displayed. Selecting Custom displays additional fields allowing you to indicate a Start Date and time and an End Date and time.

Use the Minimum Required Response Time Dashboard Data Points to configure the minimum amount of data ExtremeCloud IQ Site Engine requires before displaying a given application or site pair. The data below this threshold is not reliable and can set off a false alarm, however, you can adjust how much data is required based on the individual needs of your network.

The Network Service Dashboard displays the performance (in response time) of your network services. Each column in the dashboard represents a service:

  • ALL
  • DHCP
  • DNS
  • Kerberos
  • LDAP
  • RADIUS

The top graphs for each service displays the average response time of all of the sites for that service, while the following rows indicate the top worst performing sites for that service.

You can display or hide any of the application columns using the Views drop-down list. You can also select the X at the top of a column to hide the column from the dashboard. Select the Single Row icon () to display all columns in a single row, or select the Double Row icon () to display the columns in two rows.

The worst performing sites are defined as those whose response time is the slowest when compared to the expected response time observed over the selected time period. For example, a site with an average RADIUS authentication response time of 40 ms over the past seven days that displayed a slowest response time of 50 ms would rank as a better performing site than a site with an average RADIUS authentication response time of 5 ms over the same period that displayed a slowest response time of 30 ms.

Expected Response Time

The Expected Response Time bar graph displays the range of response times, the most recently measured response time, and the expected response time for a network service a specific site during the date range you configure in the Date Range drop-down list. The value displayed on the far right of the graph is the slowest response time observed during the selected time period. The vertical green bar indicates the most recently observed response time for the network service.

Expected Response Time bar graph

Hover over the Expected Response Time graph to display a pop-up with the response time for the network service as well as the date and time the measurement occurred. The Expected Response Time bar graphs also display the client count, represented by a number and a monitor icon ( ), and a client byte count observed as of the most recent measured minute. The client count is the number of clients using the service at the site. The client byte count indicates the amount of storage being utilized by clients. The data used for the client count, the client byte count, and the reported response time are from the same recently observed minute.

  NOTE:

Client counts and client byte counts are not provided for the bar graphs that display the average response time of all the sites for that service.

ExtremeCloud IQ Site Engine uses a standard deviation of the values gathered as response times to determine the expected response time for a network service at a site. In the bar graph, the medium gray color indicates a response time that falls within the "expected" range. A response time in the light gray range is better than expected, while a response time in the dark gray is worse than expected.

When a response time is determined to be worse than expected, the site name and the response time indicator turn red to flag the service.

Selecting the Expected Response Time bar graph opens the Response Time dashboard (which is also accessible from the Analytics > Dashboard tab) filtered to display the network service. If you select the network service for a particular site, the Response Time dashboard also filters to that site.

Historical Response Time

The Historical Response Time line graph shows all of the response times observed for the network service at a site.

Historical Response Time line graph

Hovering over a point in the graph causes a dot on the line graph to appear, indicating the point in the response time at which you are looking. Additionally, a pop-up with the date, time, and response time displays for that point.

This is the data set from which ExtremeCloud IQ Site Engine creates the Expected Response Time graph. The wider the expected response time range in the Expected Response Time graph (indicated by the medium gray color), the greater the variance in the values in this graph.


For information on related help topics: