Github user superbobry commented on the issue: https://github.com/apache/spark/pull/17230 Here're the result of piping [The Sonnets by W. Sheakespear](https://www.gutenberg.org/cache/epub/1105/pg1105.txt) via `cat` using either `PipedRDD` or hand-patched `PipedRDD` with I/O formats. The version of Spark used is 1.6.3. ``` Benchmark Mode Cnt Score Error Units pipe avgt 60 12,141 ± 0,862 ms/op pipeFormatted avgt 60 12,603 ± 0,639 ms/op ``` There is a ~4% difference in average time, which however is within the standard error. I suspect we can do better at the expense of complicating the code, so I'd rather not do it in this PR. What do you think?
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org