You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@edgent.apache.org by "Victor Dogaru (JIRA)" <ji...@apache.org> on 2016/04/07 01:00:29 UTC
[jira] [Updated] (QUARKS-11) Support for custom metrics
[ https://issues.apache.org/jira/browse/QUARKS-11?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Victor Dogaru updated QUARKS-11:
--------------------------------
Description:
Currently Quarks provides the following metrics encapsulated in Oplet implementations:
* tuple counter (number of tuples since topology submission)
* tuple meter (mean throughput and moving averages).
Additional support is needed to satisfy the following usage scenarios:
* Using metrics in lambda functions: a developer writes a function using Java metrics directly.
* Define oplet which encapsulates multiple metrics
* Metric naming: a logical name is assigned to a metric when a topology is declared; name is used to access the metric at run time.
For reference the initial discussion was added under this issue: https://github.com/quarks-edge/quarks/issues/117
was:
Currently Quarks provides the following metrics encapsulated in Oplet implementations:
* tuple counter (number of tuples since topology submission)
* tuple meter (mean throughput and moving averages).
Additional support is needed to satisfy the following usage scenarios:
* Using metrics in lambda functions: a developer writes
* Define oplet which encapsulates multiple metrics
* Metric naming: a logical name is assigned to a metric when a topology is declared; name is used to access the metric at run time.
For reference the initial discussion was added under this issue: https://github.com/quarks-edge/quarks/issues/117
> Support for custom metrics
> --------------------------
>
> Key: QUARKS-11
> URL: https://issues.apache.org/jira/browse/QUARKS-11
> Project: Quarks
> Issue Type: New Feature
> Components: Runtime
> Reporter: Victor Dogaru
> Assignee: Victor Dogaru
>
> Currently Quarks provides the following metrics encapsulated in Oplet implementations:
> * tuple counter (number of tuples since topology submission)
> * tuple meter (mean throughput and moving averages).
> Additional support is needed to satisfy the following usage scenarios:
> * Using metrics in lambda functions: a developer writes a function using Java metrics directly.
> * Define oplet which encapsulates multiple metrics
> * Metric naming: a logical name is assigned to a metric when a topology is declared; name is used to access the metric at run time.
> For reference the initial discussion was added under this issue: https://github.com/quarks-edge/quarks/issues/117
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)