Hi Pat, I’m still trying to understand the implications of Java’s Class Hierarchy Analysis [0].
Flink currently uses only a single implementation of InMemorySorter, which is NormalizedKeySorter. FLINK-4705 adds support for FixedLengthRecordSorter for Flink’s Value types and Tuples. This proposal compiles an additional implementation of InMemorySorter for each key length. The calls from QuickSort and HeapSort will now be polymorphic during sorting, as will the calls from the drivers and UnilateralSortMerger. Greg [0]: http://flink.apache.org/news/2015/09/16/off-heap-memory.html <http://flink.apache.org/news/2015/09/16/off-heap-memory.html> [1]: https://issues.apache.org/jira/browse/FLINK-4705 <https://issues.apache.org/jira/browse/FLINK-4705> > On Mar 7, 2017, at 10:06 AM, pat.chormai <pat.chor...@gmail.com> wrote: > > Hi all, > > We have almost finished implementing the functionalities. The implementation > is available at [1]. > > Also, we have included the benchmark result of FLINK-3722 into the FLIP[2] > as well as other implementation details > > We would be appreciated if you could give us feedback on this before > actually submitting PR to Flink repo. > > Best, > Pat > > [1] https://github.com/heytitle/flink/pull/1 > [2] > https://docs.google.com/document/d/1anGQhBn9qI0yqe7twVvrDIiym4U4gxalJkZzM4Ar4QM > > > > -- > View this message in context: > http://apache-flink-mailing-list-archive.1008284.n3.nabble.com/FLINK-5734-Code-Generation-for-NormalizedKeySorter-tp15804p16377.html > Sent from the Apache Flink Mailing List archive. mailing list archive at > Nabble.com.
