flink-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From zentol <...@git.apache.org>
Subject [GitHub] flink pull request #2386: [FLINK-3660] Measure latency and exposes them via ...
Date Thu, 08 Sep 2016 12:32:48 GMT
Github user zentol commented on a diff in the pull request:

    https://github.com/apache/flink/pull/2386#discussion_r77996178
  
    --- Diff: docs/monitoring/metrics.md ---
    @@ -475,52 +474,76 @@ Flink exposes the following system metrics:
           <td></td>
         </tr>
         <tr>
    -      <tr>
    -        <th rowspan="7"><strong>Task</strong></t>
    -        <td>currentLowWatermark</td>
    -        <td>The lowest watermark a task has received.</td>
    -      </tr>
    -      <tr>
    -        <td>lastCheckpointDuration</td>
    -        <td>The time it took to complete the last checkpoint.</td>
    -      </tr>
    -      <tr>
    -        <td>lastCheckpointSize</td>
    -        <td>The total size of the last checkpoint.</td>
    -      </tr>
    -      <tr>
    -        <td>restartingTime</td>
    -        <td>The time it took to restart the job.</td>
    -      </tr>
    -      <tr>
    -        <td>numBytesInLocal</td>
    -        <td>The total number of bytes this task has read from a local source.</td>
    -      </tr>
    -      <tr>
    -        <td>numBytesInRemote</td>
    -        <td>The total number of bytes this task has read from a remote source.</td>
    -      </tr>
    -      <tr>
    -        <td>numBytesOut</td>
    -        <td>The total number of bytes this task has emitted.</td>
    -      </tr>
    -    </tr>
    -    <tr>
    -      <tr>
    -        <th rowspan="3"><strong>Operator</strong></th>
    -        <td>numRecordsIn</td>
    -        <td>The total number of records this operator has received.</td>
    -      </tr>
    -      <tr>
    -        <td>numRecordsOut</td>
    -        <td>The total number of records this operator has emitted.</td>
    -      </tr>
    -      <tr>
    -        <td>numSplitsProcessed</td>
    -        <td>The total number of InputSplits this data source has processed.</td>
    -      </tr>
    +      <th rowspan="7"><strong>Task</strong></th>
    +      <td>currentLowWatermark</td>
    +      <td>The lowest watermark a task has received.</td>
    +    </tr>
    +    <tr>
    +      <td>lastCheckpointDuration</td>
    +      <td>The time it took to complete the last checkpoint.</td>
    +    </tr>
    +    <tr>
    +      <td>lastCheckpointSize</td>
    +      <td>The total size of the last checkpoint.</td>
    +    </tr>
    +    <tr>
    +      <td>restartingTime</td>
    +      <td>The time it took to restart the job.</td>
    +    </tr>
    +    <tr>
    +      <td>numBytesInLocal</td>
    +      <td>The total number of bytes this task has read from a local source.</td>
    +    </tr>
    +    <tr>
    +      <td>numBytesInRemote</td>
    +      <td>The total number of bytes this task has read from a remote source.</td>
    +    </tr>
    +    <tr>
    +      <td>numBytesOut</td>
    +      <td>The total number of bytes this task has emitted.</td>
    +    </tr>
    +    <tr>
    +      <th rowspan="4"><strong>Operator</strong></th>
    +      <td>numRecordsIn</td>
    +      <td>The total number of records this operator has received.</td>
    +    </tr>
    +    <tr>
    +      <td>numRecordsOut</td>
    +      <td>The total number of records this operator has emitted.</td>
    +    </tr>
    +    <tr>
    +      <td>numSplitsProcessed</td>
    +      <td>The total number of InputSplits this data source has processed (if the
operator is a data source).</td>
    +    </tr>
    +    <tr>
    +      <td>latency</td>
    +      <td>A latency gauge reporting the latency distribution from the different
sources.</td>
         </tr>
       </tbody>
     </table>
     
    +
    +### Latency tracking
    +
    +Flink allows to track the latency of records traveling through the system. To enable
the latency tracking
    +a `latencyTrackingInterval` (in milliseconds) has to be set to a positive value in the
`ExecutionConfig`.
    +
    +At the `latencyTrackingInterval`, the sources will periodically emit a special record,
called a `LatencyMaker`.
    +The marker contains a timestamp from the time when the record has been emitted at the
sources.
    +Latency marker can not overtake regular user records, thus if records are queuing up
in front of an operator, 
    --- End diff --
    
    marker -> markers


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

Mime
View raw message