Elias Bassani created SPARK-19124:
--------------------------------------
Summary: GraphX PageRank execution time
Key: SPARK-19124
URL: https://issues.apache.org/jira/browse/SPARK-19124
Project: Spark
Issue Type: Question
Components: GraphX, Spark Shell
Affects Versions: 2.0.2
Reporter: Elias Bassani
Hello, I don't know if I'm writing in the right place but if anyone can help me
that would be great.
I've to run PageRank on a really big graph, 400 million edges, 12 million
vertices (Wikipedia's graph) but It raises an execution time problem: after 10+
iteration of the algorithm the execution time raises abnormally from 10 mins
per iteration to dozens of hours: https://d.pr/svBR.
My code is really simple and it's taken directly from GraphX documentation.
The machine used has two CPU Intel Xeon E5-2697 v3, 64GB of RAM and 500GB hard
disk and it runs Windows Server 2012 R2 Standard.
I allocated 8 cores and 50 GB of RAM to Spark invoking the Spark-Shell from the
command line.
What could the problem be?
Thanks for any help!
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
