Re: [DISCUSS] FLIP-18: Code Generation for improving sorting performance

2017-04-06 Thread Pattarawat Chormai
7e909327d28f >> [2]https://docs.google.com/spreadsheets/d/1PcdCdFX4bGecO6Lb5dLI2nww2NoeaA8c3MgbEdsVmk0/edit#gid=598217386 >> >> Best, >> Pat >> >> >> >> -- >> View this message in context: >> http://apache-flink-mailing-list-archive.1008284.n3.nabble.com/DISCUSS-FLIP-18-Code-Generation-for-improving-sorting-performance-tp16486p16923.html >> Sent from the Apache Flink Mailing List archive. mailing list archive at >> Nabble.com. >

Re: [DISCUSS] FLIP-18: Code Generation for improving sorting performance

2017-04-05 Thread Greg Hogan
/1PcdCdFX4bGecO6Lb5dLI2nww2NoeaA8c3MgbEdsVmk0/edit#gid=598217386 > > Best, > Pat > > > > -- > View this message in context: > http://apache-flink-mailing-list-archive.1008284.n3.nabble.com/DISCUSS-FLIP-18-Code-Generation-for-improving-sorting-performance-tp16486p16

Re: [DISCUSS] FLIP-18: Code Generation for improving sorting performance

2017-04-03 Thread Pattarawat Chormai
#gid=598217386 Best, Pat -- View this message in context: http://apache-flink-mailing-list-archive.1008284.n3.nabble.com/DISCUSS-FLIP-18-Code-Generation-for-improving-sorting-performance-tp16486p16923.html Sent from the Apache Flink Mailing List archive. mailing list archive at Nabble.com.

Re: [DISCUSS] FLIP-18: Code Generation for improving sorting performance

2017-03-24 Thread Pattarawat Chormai
Hi, Gábor and I plan to investigate the metamorphic call issue further. We will implement the idea of combining QuickSort and NormalizedKeySorter together in generated code and benchmark the improvement with a Flink’s job that uses 3 different sorters. Best, Pat > On Mar 23, 2017, at 6:31 PM

Re: [DISCUSS] FLIP-18: Code Generation for improving sorting performance

2017-03-23 Thread Gábor Gévay
Hello, > Second, additional classes will turn performance critical callsites > megamorphic. Yes, this is a completely valid point, thanks for raising this issue Greg. We were planning to have an offline discussion tomorrow with Pattarawat about this. We have a few options: 1. We could fuse the Q

Re: [DISCUSS] FLIP-18: Code Generation for improving sorting performance

2017-03-23 Thread Greg Hogan
I would be more than happy to shepherd and review this PR. I have two discussion points. First, a strategy for developing with templates. IntelliJ has a FreeMarker plugin but we lose formatting and code completion. To minimize this issue we can retain the untemplated code in an abstract class w

Re: [DISCUSS] FLIP-18: Code Generation for improving sorting performance

2017-03-21 Thread Fabian Hueske
Hi Pat, thanks a lot for this great proposal! I think it is very well structured and has the right level of detail. The improvements of your performance benchmarks look very promising and I think code-gen'd sorters would be a very nice improvement. I like that you plan to add a switch to activate