phaniarnab opened a new pull request, #1777: URL: https://github.com/apache/systemds/pull/1777
This patch enables reuse of RDDs of redundant Spark operations. We also persist a subset of operations in the executors, where the rest are just cached locally. Reuse of even unpersisted RDDs allows Spark to apply optimizations and skip stages. In addition, this patch removes the compile-time flag to indicate reuse and instead reuse all RDDs. Local RDD caching is now disabled due to bugs. LinCache Spark (Col/Loc/Dist): 16/2/2. => indicates the number of reused collects/prefetches (=16), local RDDs (=2) and persisted RDDs(=2). -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
