Download the Rev Connect Zone Report
Event Hosts and Moderators may access details about all Rev Connect zones once an event has ended.
Note: If you have not enabled Rev Connect zones for your Webcasts this report will not be visible for download. |
To access the Rev Connect zone report:1. Navigate to Event Calendar > Event Name.
2. Click Reports.(Note: You may need to specify which reports you want to view. Main Event or Pre-Production Event may be specified if you run each type of Webcast.).
3. Click Download > Rev Connect - CSV.
Each line within the Rev Connect report represents one user within a Peer Mesh. In most cases, users join a single Peer Mesh for an entire event and, therefore, there is only one entry. However, if a user joins or rejoins one or more Peer Meshes, a line is added for each instance. This provides different user experiences based on membership in different Peer Meshes available to the analyst during review. The UserID should be used for grouping information about specific Users – and UserID is the same and used for correlation to the Webcast Event Attendees Report.
If viewed as a table (as in Excel), the first set of columns focus on the User (or viewer) characteristics. They include:
●Attendee Full Name (First Name / Last Name)
●Account Username
●UserID
●Email Address
●Session ID
●Zone Name
●IP Address
●Browser Used
●Origin URL
Columns after User characteristics focus on Peer Mesh data. They include:
●Peer Mesh GUID
○The Peer Mesh GUID is an unique ID representing a Peer Mesh. During analysis, this item can be used in a number of ways to:
●Identify how many Peer Meshes were in this event (by counting unique Peer Mesh GUIDs)
●Identify the timing life cycle of the Peer Mesh with start of this Peer Mesh (earliest Time In) and the end of the Peer Mesh (latest Time Out)
●Grouping within Peer Mesh for analysis of:
●Review User IPs for validation of the Peer Mesh membership (defined within the Zone settings for Peer Mesh)
●Compare Peer Mesh metrics (defined below)
●Peer Mesh Shares Out (number of segments that this viewer provide to another peer)
○This is the number of segments that this viewer, in this particular Peer Mesh, has shared to another peer. Higher number represents more sharing. Higher numbers are not necessarily good nor bad, but should be considered if a user is sharing too much. The Samaritan Load measure below expands on this and provides a consistent way to compare different users and their sharing behavior.
●Peer Mesh Shares In (number of segments that this viewer gets from another peer)
○This is the number of segments that this viewer got from another another peer. Higher numbers represent getting more shares from peers and a better use of bandwidth.
●Origin Gets
○This is the number of segments that this viewer has gone to origin (i.e., not a peer) to retrieve. While it is important that this number be low, it should also be realized that “someone” (some viewer) will have to go to origin to get it. This should always be evaluated in reference to Peer Mesh Shares Out, Peer Mesh Shares In and Total Segments. Having one (or a few) peers bear the load of many Origin Gets is not necessarily indicative of a problem.
●Time In
○This is the time viewer enters this Peer Mesh. Viewers may be assigned to different Peer Meshes across an event, so please use this value with Time Out and all analysis should look for multiple entries per user.
●Time Out
○Like Time In, This is the time viewer leaves this Peer Mesh. Viewers may be assigned to different Peer Meshes across an event, so please use this value with Time In and all analysis should look for multiple entries per user.
●Total Segments In
○This is the number of segments that the viewer requested within the peer mesh. This is the number of Peer Mesh Shares In (how many they get from other peers) + Origin Gets (this is the number of origin retrievals).
The following calculated percentages provide an insight to where (in percentage) the current user (within the peer mesh) is getting content (either via sharing from peers or going to origin), and how much user is sharing. While these are meaningful at the viewer level, they should not be used for an indication of the efficiency across the peer mesh.
●Viewer Peer Mesh Efficiency (Peer Mesh Shares In / Total Segments In) * 100
○This is a viewer specific calculated measure. It represents the ratio/percentage of Peer Mesh Shares In (number of segments this viewer gets from peers) to the Total Segments In (total number of segments played, in from peers + origin gets).
○In other words: This is the percentage of segments that a viewer got from a peer.
●The higher the value, the more that this viewer is using the peer mesh to get content. The lower the value, the more this viewer is going to origin.
●Viewer Peer Mesh Origin Load (Origin Gets / Total Segments In) * 100
○This is a viewer specific calculated measure, and the flip side of the Viewer Peer Mesh Efficiency. Meaning Viewer Peer Mesh Origin Load + Viewer Peer Mesh Efficiency = 100%.
○This number represents the ratio/percentage of
○Origin Gets (number of segments this viewer goes to origin for) to the Total Segments In (total number of segments played, in from peers + origin gets).
●In other words: This is the percentage of segments that a viewer needed to go to origin.
●The higher the value, the more this viewer is going to origin. The lower the value, the more that this viewer is using the peer mesh to get content.
●Viewer Peer Mesh Samaritan Load (Peer Mesh Shares Out / Total Segments In ) * 100
○This is a viewer specific calculated measure. It represents the ratio of how many times this user has shared with other peers to the total number of segments. This provides a quantitative measure to “How actively does this viewer share?” or “How good a Samaritan is this viewer in this Peer Mesh?”
○This number represents the ratio of Peer Mesh Shares Out (how many times this user shared to another peer) to Total Segments In (total segments that could be shared, Shares IN + Origin Gets). This measure is provided as a ratio/percentage, but can be > 100%.
○In other words: This is the percentage of how often this viewer shares to peers.
●Higher numbers mean this viewer is sharing more (supplying segments to more peers.) Lower numbers represent a lack of sharing from this peer.
The next set of calculated measures represent Peer Mesh wide characteristics. These are common/duplicated across each User within the Peer Mesh. Meaning, these measures will be the same for all Users within a particular Peer Mesh, and are provided in this manner for direct comparison to the Viewer specific measures identified above.
●Mesh Peer Mesh Share Efficiency = ( Sum(Peer Mesh Shares In) / Sum(Total Segments In))
○This is a Peer Mesh specific calculated measure. It represents the ratio of how many times all the users got segments from peers to the total number of segments for all users. This provides a quantitative measure to “Is this Peer Mesh sharing efficiently?”
○In other words: This percentage represents how often we are sharing.
●Higher numbers mean better sharing within this Peer Mesh. Target should be > 90%. Lower numbers represent a lack of sharing within this Peer Mesh.
○Do not confuse Viewer Peer Mesh Efficiency with Mesh Peer Mesh Share Efficiency. Each metric identifies scope by the first word, either Viewer (the single viewer) or Mesh (the single Peer Mesh) scope.
●Mesh Peer Mesh Origin Efficiency = ( Sum(Origin Gets) / Unique (Origin Gets) )
○This is a Peer Mesh specific calculated measure. It represents the ratio of how many times (across all the users) there were origin retrievals for a segment to the number of unique segments. This provides a quantitative measure to “Is this Peer Mesh going to origin too much?”
○In other words: This ratio, always greater than 100, represents the number of times we go to origin to get segments against all unique segments.
●For example, if the metric is 100 – that means we only ever go to origin once per segment (i.e., All Origin Gets == Unique Origin Gets). This implies that the users share that segment optimally between each other within the Peer Mesh (because, otherwise, there would be additional Origin Gets). 100 is the optimal result, but rarely achievable due to inherent timing between communicating peers.
●If the metric is 125, this equates to, on average, going to origin 5 times for 4 unique segments.
●Guidance: Strive for Mesh Peer Mesh Origin Efficiency value of 100-125.
●Are higher values bad? Higher values indicate more origin gets across all users, and hence more bandwidth used. High values of this metric are most impactful on Peer Meshes that exist on networks with reduced or restricted bandwidth. Networks with additional bandwidth can service more origin calls on average.
●How do I effect change on Mesh Peer Mesh Origin Efficiency? There are multiple reasons why this metric could be driven up, and if most often unique per network. A primary cause is network induced latency between Peer Mesh constituents/viewers. A large latency between communicating peers will time-out resulting in more requests driven to origin.
●Examine your Rev Connect Peer membership rules on the Zones page. Only allow Peer Meshes to be comprised of “close” (restricted to a small geographical/network grouping) viewers.
●Consider the size and structure of your HLS stream. Smaller segment sizes, while reducing latency, increase the communication load across all peers and the network. Larger sizes allow for more time to communicate between peers.
○Do not confuse Viewer Peer Mesh Origin Load with Mesh Peer Mesh Origin Efficiency. Each metric identifies scope by the first word, either Viewer (the single viewer) or Mesh (the single Peer Mesh) scope.
Role(s): Event Host, Account Admin, Event Moderators |