[
https://issues.apache.org/jira/browse/SPARK-21057?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Apache Spark reassigned SPARK-21057:
------------------------------------
Assignee: (was: Apache Spark)
> Do not use a PascalDistribution in countApprox
> ----------------------------------------------
>
> Key: SPARK-21057
> URL: https://issues.apache.org/jira/browse/SPARK-21057
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
> Affects Versions: 2.1.1
> Reporter: Lovasoa
>
> I was reading the source of Spark, and found this:
> https://github.com/apache/spark/blob/v2.1.1/core/src/main/scala/org/apache/spark/partial/CountEvaluator.scala#L50-L72
> This is the function that estimates the probability distribution of the total
> count of elements in an RDD given the count of only some partitions.
> This function does a strange thing: when the number of elements counted so
> far is less than 10 000, it models the total count with a negative binomial
> (Pascal) law, else, it models it with a Poisson law.
> Modeling our number of uncounted elements with a negative binomial law is
> like saying that we ran over elements, counting only some, and stopping after
> having counted a given number of elements.
> But this does not model what really happened. Our counting was limited in
> time, not in number of counted elements, and we can't count only some of the
> elements in a partition.
> I propose to use the Poisson distribution in every case, as it can be
> justified under the hypothesis that the number of elements in each partition
> is independent and follows a Poisson law.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
