[
https://issues.apache.org/jira/browse/YARN-3901?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14746352#comment-14746352
]
Li Lu commented on YARN-3901:
-----------------------------
Thanks [~sjlee0]!
bq. It's essentially using AtomicLong.compareAndSet(). The few lines around it
are mostly to keep pace with the current time. I hope that makes sense.
Yes, the algorithm makes sense. I thought JVM may have some special
optimization with the getAndAdd method on x86 since x86 has a specific
fetch-and-add (LOCK:XADD). However I'm not sure if this is actually reflected
in most current JVMs (https://bugs.openjdk.java.net/browse/JDK-6973482). I'm OK
with both, since the advantage of either one of them in our use case is
nontrivial (https://blogs.oracle.com/dave/entry/atomic_fetch_and_add_vs). If
keeping up with current time is important, we need to stick to the CAS based
solution and _not_ to change it in future.
bq. We create a new record in this table any time a new activity is done for a
given day for a flow.
Thanks. I missed the {{getTopOfTheDayTimestamp}} part.
bq. The prefix is just cluster! and we can grab from the beginning. What you
mention would get activities for today only, which is slightly different.
Right. Thanks for the clarification! I meant activities for the past 24 hours.
> Populate flow run data in the flow_run & flow activity tables
> -------------------------------------------------------------
>
> Key: YARN-3901
> URL: https://issues.apache.org/jira/browse/YARN-3901
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: timelineserver
> Reporter: Vrushali C
> Assignee: Vrushali C
> Attachments: YARN-3901-YARN-2928.1.patch,
> YARN-3901-YARN-2928.2.patch, YARN-3901-YARN-2928.3.patch,
> YARN-3901-YARN-2928.4.patch, YARN-3901-YARN-2928.5.patch,
> YARN-3901-YARN-2928.6.patch, YARN-3901-YARN-2928.7.patch,
> YARN-3901-YARN-2928.8.patch
>
>
> As per the schema proposed in YARN-3815 in
> https://issues.apache.org/jira/secure/attachment/12743391/hbase-schema-proposal-for-aggregation.pdf
> filing jira to track creation and population of data in the flow run table.
> Some points that are being considered:
> - Stores per flow run information aggregated across applications, flow version
> RM’s collector writes to on app creation and app completion
> - Per App collector writes to it for metric updates at a slower frequency
> than the metric updates to application table
> primary key: cluster ! user ! flow ! flow run id
> - Only the latest version of flow-level aggregated metrics will be kept, even
> if the entity and application level keep a timeseries.
> - The running_apps column will be incremented on app creation, and
> decremented on app completion.
> - For min_start_time the RM writer will simply write a value with the tag for
> the applicationId. A coprocessor will return the min value of all written
> values. -
> - Upon flush and compactions, the min value between all the cells of this
> column will be written to the cell without any tag (empty tag) and all the
> other cells will be discarded.
> - Ditto for the max_end_time, but then the max will be kept.
> - Tags are represented as #type:value. The type can be not set (0), or can
> indicate running (1) or complete (2). In those cases (for metrics) only
> complete app metrics are collapsed on compaction.
> - The m! values are aggregated (summed) upon read. Only when applications are
> completed (indicated by tag type 2) can the values be collapsed.
> - The application ids that have completed and been aggregated into the flow
> numbers are retained in a separate column for historical tracking: we don’t
> want to re-aggregate for those upon replay
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
