6 ms·
Loop Recognition Benchmarks in C++, Java, Go, Scala (2011) [pdf]
- kasey_junk 12y agoNeeds a 2011 tag.
- pella 12y agoyes - 2011 http://research.google.com/pubs/pub37122.html http://research.google.com/pubs/pub37122.html Venue : Proceedings of Scala Days 2011 Publication Year: 2011
- dang 12y agoYes, and the title was rewritten as well [1]. Submitters: please use an article's title unless it is linkbait or misleading [2]. 1. Submitted title was "Google Publishes C++, Go, Java and Scala Performance Benchmarks" 2. https://news.ycombinator.com/newsguidelines.html https://news.ycombinator.com/newsguidelines.html
- barrkel 12y agoIMO the title is misleading, and the submitted title is better. Though "Google Researcher Publishes..." would more accurately state its provenance. The algorithm encoded in the benchmark is loop recognition, but the purpose of the paper is to benchmark the languages, not to find loops.
- dang 12y agoOk, thanks. We added "Benchmarks" to the title.
- saltcookie 12y agohttp://readwrite.com/2011/06/06/cpp-go-java-scala-performance-benchmark http://readwrite.com/2011/06/06/cpp-go-java-scala-performanc... this is the real artical, which just points to the pdf
- saltcookie 12y agoThat is true, i don't understand y some one changed the tittle
- gtani 12y agoand this phrase on page 7 highlighted The benchmarking itself was done in a simple and fairly un-scientific fashion. ... older Pentium IV workstation. Run-times were measured using wall-clock time. but still a good read about tweaking c++ (I think at the time, clang and gcc were still implementing the C++11 standard features). Also i remember long HN threads about this in mid 2011 but i can't find them, here'a a later discussion https://news.ycombinator.com/item?id=4539668 https://news.ycombinator.com/item?id=4539668
- f2f 12y agothe go side of that paper should be accompanied by this link, always: http://blog.golang.org/profiling-go-programs http://blog.golang.org/profiling-go-programs
- pella 12y agobenchmark - now 2014 : http://www.techempower.com/benchmarks/ http://www.techempower.com/benchmarks/
- broodbucket 12y agoWhile this tells a somewhat similar story, comparing web frameworks to the languages themselves is apples to oranges.
- realrocker 12y agoThis is really old. C++, Java are mature languages. Scala hit 2.11 in March 2014, while Go version 1 was released in March 2012 and currently is at 1.3. Versions don't tell the whole story, but they do matter.
- claudius 12y agoOn the other hand, this: // Step d: IntList::iterator back_pred_iter = back_preds[w].begin(); IntList::iterator back_pred_end = back_preds[w].end(); for (; back_pred_iter != back_pred_end; back_pred_iter++) { int v = *back_pred_iter; … can now be written for (auto v : back_preds[w]) { … , hence you could argue that while C++ is certainly “mature” in a sense, it is far from converged.
- dcsommer 12y ago> Go and Scala have powerful type inference, making explicit type declarations very rare. In C++ and Java everything needs to be declared explicitly. I guess they forgot about C++11's auto?
- masklinn 12y agoAnd "powerful type inference"? For Go? Doesn't it just do local implicit typing using the RHS type straight as the LHS's?
- igouy 12y agoDiscussion from 2011 https://news.ycombinator.com/item?id=2615096 https://news.ycombinator.com/item?id=2615096