Skip to content

Dashboard

The Edge Services Platform provides you with an Edge Analytics dashboard that gives you instant visibility on key metrics across the entire streaming delivery chain. The added observability allows you to easily understand the state of the underlying infrastructure performance, audience, statistics, among others.The dashboard offers the ability to analyze specific events and/or filter metrics by a number of specific attributes.

All this data can help you plan and optimize the content for your specific needs.

Edge Analytics Dashboard

The dashboard offers a different set of charts based on the Analytics tier associated with your account.

Note

You can also use a Self-hosted Grafana integrated with Edge Analytics

Analytics tiers

The base configuration is the Lite Analytics tier, which is available when you create your account and can be upgraded to the Advanced tier. All the features of the Lite tier are also available on the Advanced tier.

Dashboard on the Lite tier

platform-dashboard-lite
Platform dashboard with Lite Analytics tier

Dashboard on the Advanced tier

platform-dashboard-advanced
Platform dashboard with Advanced Analytics tier

Note

Please refer to sales@system73.com for any inquiry on contracting the Advanced tier.

Charts

The dashboard is grouped into sections for convenience.

KPI and summary

The first section provides you with key performance indicators such as:

Peak audience

The highest number of devices connected simultaneously for the selected time period.

Source: SQL endpoint /analytics/v1/query/sql

CDN data

The total data transferred over CDN for the selected time period.

Source: SQL endpoint /analytics/v1/query/sql

P2P data

The total data transferred over P2P for the selected time period.

Source: SQL endpoint /analytics/v1/query/sql

P2P Offload

The total P2P offload percentage of the video distribution for the selected time period.

Source: SQL endpoint /analytics/v1/query/sql

Join time

The median of the amount of time that takes from loading the connection to the first video frame.

Source: SQL endpoint /analytics/v1/query/sql

Total viewing time

The total of playtime for the selected time period.

Source: SQL endpoint /analytics/v1/query/sql

Summary table

Provides a fine grain view of the mentioned KPIs as well as filter capabilities.

summary-table
Summary table

Source: SQL endpoint /analytics/v1/query/sql

Audience engagement

The Engagement section displays information regarding the engagement of your audience.

Audience

Number of playback sessions within the selected time range, aggregated per time granularity interval.

Source: SQL endpoint /analytics/v1/query/sql

Optimal viewport resolution

The total number of devices obtaining the best resolution they can represent.

Source: SQL endpoint /analytics/v1/query/sql

Suboptimal viewport resolution

The total number of devices obtaining lower resolution than they can represent.

Source: SQL endpoint /analytics/v1/query/sql

Join time

The time it takes to present the first frame to the viewer after the service finishes loading. Bars are distributed by four quantiles:

  • Q1 (25th percentile): This is the value below which 25% of the data falls.
  • Median (50th percentile): This is the middle value of the dataset, where 50% of the data falls below and 50% above.
  • Q3 (75th percentile): This is the value below which 75% of the data falls.
  • 97.5th percentile: This indicates the value below which 97.5% of the data falls.

join-time.png

Source: SQL endpoint /analytics/v1/query/sql

Optimal viewport percent

Percentage of total playing viewers at an optimal/suboptimal viewport resolution, where the "optimal" corresponds to the audience consuming the video representation with a resolution closest to the viewport reported by the video player.

Source: Metrics endpoint /analytics/v1/metrics/qoe/optimal_viewport/percentage

Player state

Percentage of the reported player sessions within the selected time range, aggregated by player state.

  • starting: the starting state measures the initial buffering and its related player internal states.
  • buffering: the buffering state measures buffering events that impact user experience, preventing video playback.
  • playing: the accumulated time in the life cycle of the SDK when the player state is playing. This is the total time watching video within the viewing session.
  • pause: the accumulated time in the life cycle of the SDK when the player state is paused. This is the total time with video paused within the viewing session.
  • unknown: the player is in an unknown state, such as when the player has not attached a stream, has produced an error, has ended the stream, the integrator didn't pass a value, or passed an incorrect value, or similar. This state is used when the other states are not applicable.

Source: SQL endpoint /analytics/v1/query/sql

Viewed hours

The total number of hours the content is being played over time, per granularity time window.

Advanced tier only

Source: SQL endpoint /analytics/v1/query/sql

Minutes per play

The average playtime that the content has been consumed by the audience per session over time.

Advanced tier only

Source: SQL endpoint /analytics/v1/query/sql

Buffering % (< 10 min in Buffering)

Percentage of BufferingTime during PlayingTime for Devices that have been less than 10 min in a buffering state.

Advanced tier only

Source: SQL endpoint /analytics/v1/query/sql

Traffic

Displays information of the traffic of your content distribution.

P2P Offload and Data Traffic

Peer to Peer offload and all the data transferred over CDN (Content Delivery Network/s) and P2P (Peer to Peer).

data-traffic-p2p-offload.png

Source: SQL endpoint /analytics/v1/query/sql

Connection speed distribution

Connection speed distribution within the selected time range.

Source: SQL endpoint /analytics/v1/query/sql

Connection speed

Total audience distribution by connection speed.

Source: SQL endpoint /analytics/v1/query/sql

CDN Traffic per Edge Delivery CDN Label

All the data transferred over the Content Delivery Network pathway(s) configured on System73 Edge Delivery service.

Source: SQL endpoint /analytics/v1/query/sql

CDN Traffic per X-CDN Header

HTTP request from the CDN should be included in the header in order to be available on the chart.

Source: SQL endpoint /analytics/v1/query/sql

Browsers

Total audience distribution by Browser for web platforms.

Advanced tier only

Source: SQL endpoint /analytics/v1/query/sql

Operating Systems

Total audience distribution by Operating Systems.

Advanced tier only

Source: SQL endpoint /analytics/v1/query/sql

Players & Apps

Total audience distribution by native apps or web media players.

Advanced tier only

Source: SQL endpoint /analytics/v1/query/sql

Connection types

Total audience distribution by their connection type.

Advanced tier only

Source: SQL endpoint /analytics/v1/query/sql

ASN traffic

This section is only available on the Advanced Analytics tier and displays traffic of your content distribution per each ASN.

Audience by ASN

Total audience distribution by their ASN (showing separately 'Top 5' + Others).

Source: Metrics endpoint /analytics/v1/metrics/audience/asn

Traffic by ASN: Volume

Total amount of CDN traffic incoming from each ASN (showing separately 'Top 5' + Others).

Source: Metrics endpoint /analytics/v1/metrics/traffic/asn

Traffic by ASN: Percentage

Total traffic volume in percentages from each ASN (showing separately 'Top 5' + Others).

Source: Metrics endpoint /analytics/v1/metrics/traffic/asn/percentage

Geographical maps

Displays the audience distribution on geographical maps.

Geo map

Provides latitude and longitude points distribution of the audience.

Advanced tier

  • Geo map: On the advanced tier the points distribution of latitude and longitude also provides the ASN information.

Source: SQL endpoint /analytics/v1/query/sql

Geo heatmap

Provides a heatmap of the audience distribution over the world map.

Source: SQL endpoint /analytics/v1/query/sql

Network topology

3D Globe

Provides a geospatial distribution of the audience through a 3D Globe graph. The user can interact with it by rotating the globe and update the information shown by changing the timeframe and other dashboard filter options.

3D Globe Chart Overview

3d-globe-preview
3D Globe initial rendering

Note

Please note that data points from the past are limited in order not to overload the chart. You can adjust the visible timeframe by changing the "To" date in dashboard filters.

Timeframe info notification

3d-globe-tooltip
Info icon and tooltip are shown to identify past data points limitation

Source: SQL endpoint /analytics/v1/query/sql

Audience Topology

Provides the network topology of the audience, showcasing their origin and position from the CDN and allowing the user to explore individual node information based on their Analytics tier account. Please note that the available data retention is up to last 30 days.

3D Topology Chart Overview

3d-topology-overview
3D Topology rendering, the controls panel and general info slots placed at the top left and top right respectively

This chart contains the following features:

  • Metric status: From the controls panel, it allows selecting a metric (Bufferhealth, AFBR, or NatType) and coloring the nodes to identify their status over the selection.

Advanced tier

  • This control is only available on Advanced tier.
  • Layout: It applies a layout to the graph customizing the nodes directionality. You can choose from Top-Down (it gives a tree form to the chart) or Globular (it does not give any direction to the chart and it is spread horizontally).

  • Show only P2P nodes: This option allows you to unload/load the chart by hidden or showing nodes that are not P2P.

  • Nodes threshold control: Due to performance the tree is limited to render a fixed number of nodes (5K) and you can change this limit by setting an amount through the slider control. Please note that only it will appear when nodes results surpass the number suggested.

  • General results information: On the top right side there are a couple of slots showing the timestamp and total number of nodes loaded from datasource according to the filters selected.

Note

  • The timestamp shown is the point in time used to represent the audience within the selected time range.
  • Collapsed nodes groups: To bring a better user experience, it is implemented nodes groups that collapse other nodes along the tree with an indicator of the nodes number contained; you can expand them to see in detail the nodes information previously collapsed. Please note that this feature is only available if the total number of nodes is greater than the fixed limit suggested.
Group nodes and expanding reference

3d-topology-groups-expanding-animation
Groups rendering and expanding action reference

  • Chart interactivity details: The chart is adapted to let you explore nodes information by rotating the tree diagram, hovering the nodes to display individual information of each one, seeing groups number indicator when you are approaching by zooming into them, letting you focus on the nodes expanded approaching the point view to them.

  • Other internal improvements: The chart balances the form to render properly the nodes avoiding their collision, keeping a good image quality and performance.

Source: SQL endpoint /analytics/v1/query/sql

Analytics tier information

Based on your Analytics tier account, you are able to check some nodes information items:

  • Session Id
  • Node Id
  • AFBR metric *
  • Bufferhealth metric
  • Capacity *
  • P2P Connection Status *
  • NatType metric *
  • IP *
  • ASN *

Advanced tier

  • The items highlighted with (*) are only visible on the advanced tier.
Advanced tier nodes information

3d-topology-advanced-tier-info
Advanced tier information items shown per node

Lite tier nodes information

3d-topology-lite-tier-info
Lite tier information items shown per node

Filtering

We offer you the possibility to alter the data displayed on the dashboard by applying time intervals and/or one or more filters of your choice.

filter-variables

  1. You can filter by Channel Id, Stream URI, Bitrate and Type. For those, by default no filter is applied, while the Time granularity filter must and will always have a value.
  2. On the time interval picker you are free to choose from the quick relative time ranges or craft your own absolute time intervals.
  3. You can set your dashboard to auto refresh if you wish with this setting.

Accuracy

Important

This setting is only available on the dashboard of the Advanced analytics tier.

Along with the filters you can set the accuracy of the results for charts that are based on queries that perform a count distinct aggregation.

accuracy-var

There are two modes of accuracy:

  • Fast: results are not guaranteed to be exact but the chart queries run faster.
  • Precise: results are guaranteed to be exact at the expense of queries running slower.

The performance impact is noticeable when browsing data for large time intervals (a month or more) and/or fine time granularities.

Charts that are affected by this setting:

  • Peak Audience
  • Audience
  • Optimal Viewport Resolution
  • Suboptimal Viewport Resolution
  • Optimal Viewport Percent
  • Connection Speed
  • Browsers
  • Operating Systems
  • Players & Apps
  • Connection Types
  • Geo Map
  • Geo Heatmap

Data layers

The platform dashboards are built on top of the data layers presented on our architecture section. Please note that the data displayed in the charts is retrieved from one of the aforementioned data layers depending on the time interval selected, under the following conditions:

  1. Realtime layer: Time interval within the last 30 days spanning:
    1. Up to 1 hour for second granularity.
    2. Up to 12 hours for minute granularity.
  2. Historical layers: Time interval within the last 13 months up to 1 hour ago spanning:
    1. Up to 48 hours for minute granularity.
    2. Up to 30 days for hour granularity.
    3. From more than 2 days up to 13 months for day granularity.
    4. From 30 days up to 13 months for week and month granularities.

Note

Take into account that the time picker takes into account the DST changes so when you select the Last X days and within the last X days a DST change occurred the calculated time interval could be X-1 days and 23 hours to account for the +/- 1 hour shift.

To ignore DST changes you can also set the UTC timezone on the time picker.


This section was last updated 2024-09-17