spark-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Qi Song <songqi1...@gmail.com>
Subject Re: Are there any plans to develop Graphx Streaming?
Date Mon, 21 Apr 2014 01:27:13 GMT
Hi~Ankurdave~
Now I get another question, I realized that GraphX provides four different
graph partition methods: RandonVertexCut, CanonicalRandomVertexCut,
EdgePartition1D and EdgePartition2D. I've test the running time of these
four method using pagerank in several different datasets and found that
RandomVertexCut is not always the fastest method. I think that why GraphX
provides four different methods. But I wander if there exists some
documentation about how to choose partition methods, based on the graph's
structure or some other properties?

Thanks~



--
View this message in context: http://apache-spark-user-list.1001560.n3.nabble.com/Are-there-any-plans-to-develop-Graphx-Streaming-tp2689p4512.html
Sent from the Apache Spark User List mailing list archive at Nabble.com.

Mime
View raw message