Recorder Analytics Service Engines
The Recorder Analytics Service Engines object contains counters for monitoring the flow of data through the message queues that exist between the Engine and the Service.
-
The metadata queue used to provide data to an Engine is simpler and so is not as highly instrumented as the queue used for results data, which is of the same type as the queues used for Provider data.
-
Each instance is named after the internal name of the engine.
The following counters are available in the Recorder Analytics Service Engines object.
Name |
Description |
Notes |
---|---|---|
Alert XML Updates |
Number of alert update results received from Engine |
Tracks updates to alert XML. These results build up the XML used by the Analytics Service when required by Analytics Rules to: Fire Alert notifications Send External URL HTTP POST messages. |
Alert XML Updates/sec |
Rate of receipt of alert XML update messages from Engine |
Rate version of the previous counter. |
Bookmark Event Results |
Number of event results received from Engine |
Tracks bookmark events; these results are the items visible in the Tags panel, available when replaying recorded interactions. |
Bookmark Event Results/sec |
Rate of receipt of bookmark event messages from Engine |
Rate version of the previous counter. |
Condition Updates |
Number of condition update results received from Engine |
Tracks messages indicating that a single Analytics Rule Condition has been matched. |
Condition Updates/sec |
Rate of receipt of Condition update messages from Engine |
Rate version of the previous counter. |
Metadata Queue Size |
Number of queued metadata messages |
Tracks queued metadata messages carrying start/stop information and metadata to the Engine. |
PCM Buffer Count |
Number of PCM buffers |
Tracks PCM buffers. A 5-second PCM buffer is created for every individual audio stream. Stereo audio can be provided in one or two streams depending upon the requirements of the engine. |
PCM Buffered (ms) |
Total duration of PCM audio buffered |
Tracks the total amount of unread decompressed audio currently stored across all PCM buffers for this Engine. |
PCM Discard Rate (ms/sec) |
Rate of discarding of PCM audio data from buffers (milliseconds of audio per second) |
Rate version of the previous counter. |
PCM Discarded (ms) |
Total duration of PCM audio discarded from buffers (milliseconds) |
Tracks the total amount of PCM audio that was unread on buffers when they were discarded (due to the Engine deciding not to process any further audio without waiting for the end of the stream); total read + total discarded = total written. |
PCM Read (ms) |
Total duration of PCM read from buffers |
Tracks the total amount of PCM audio that has ever been read from PCM buffers by this Engine. |
PCM Read Rate (ms/sec) |
Rate of reading of PCM audio data from buffers |
Rate version of the previous counter. |
PCM Write Rate (ms/sec) |
Rate of writing of PCM audio data into buffers |
Rate version of the previous counter. |
PCM Written (ms) |
Total duration of PCM written to buffers |
Tracks the total amount of PCM audio that has ever been written to PCM buffers for this Engine (this counter can count the same audio twice if the Engine absorbs stereo audio as two mono streams). |
Processing Completed Notifications |
Number of processing completed messages received from Engine |
Tracks messages indicating when an Engine has completed all processing of an interaction In Speech Analytics, an interaction represents a single part of the contact between one employee and the same customer. In Text Analytics, an interaction is the communication session between one or more employees and the same customer with a unifying contextual element.. |
Processing Completed Notifications/sec |
Rate of receipt of processing completed messages from Engine |
Rate version of the previous counter. |
Results Queue Idle Thread Count |
Number of threads processing results queue currently idle |
Tracks results queue activity. Result messages provide data to the Service from the Engine; there are many possible types of messages. The queue has one subqueue per INum, ensuring that the messages for each INum are always processed in order even if there are multiple threads processing in parallel (though only one thread is ever used per queue). |
Results Queue Size |
Number of queued result messages |
|
Results Queue Subqueue Count |
Number of result subqueues |
|
Results Queue Thread Count |
Number of threads processing results queue |
|
Satellite Tag Results |
Number of Satellite tag results received from Engine |
Tracks satellite tag messages used by some Engines. For example, these messages carry the data used in the display of biometric information for interactions in Risk Management Component of the Real-Time Analytics (RTA) Framework that provides interaction search and replay functionality, combined with integration of the Real-Time Analytics (RTA) Framework. search results. |
Satellite Tag Results/sec |
Rate of receipts of satellite tag result messages from Engine |
Rate version of the previous counter. |
Tag Results |
Number of tag results received from Engine |
Tracks direct attribute tag messages produced by the Engine (rather than by an Analytics Rule Action). |
Tag Results/sec |
Rate of receipt of tag result messages from Engine |
Rate version of the previous counter. |