jmzhoulab opened a new pull request #27111: [SPARK-30441][Graphx]Improve the memory usage in StronglyConnectedComponents URL: https://github.com/apache/spark/pull/27111 ### What changes were proposed in this pull request? Make the Graph/RDD as non-persistent for 'sccGraph' in 'do ... while' and 'pregel'. The variable 'sccGraph' has been removed, use the variable 'sccVertexLabel' to collect the vertices to be removed in iteration. ### Why are the changes needed? To improve the memory usage, mark the Graph/RDD as non-persistent in the iterative process timely. figure1: before optimizing memory usage ![figure1](https://user-images.githubusercontent.com/43285458/71855828-18511100-311d-11ea-8abc-7adc669a89cb.png) figure2: after optimizing memory usage ![figure2](https://user-images.githubusercontent.com/43285458/71855848-299a1d80-311d-11ea-91a3-ee3a53125921.png) ### Does this PR introduce any user-facing change? no ### How was this patch tested? existing tests
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org