[jira] [Created] (FLINK-1983) Remove dependencies on Record APIs for Spargel

2015-05-06 Thread Henry Saputra (JIRA)
Henry Saputra created FLINK-1983: Summary: Remove dependencies on Record APIs for Spargel Key: FLINK-1983 URL: https://issues.apache.org/jira/browse/FLINK-1983 Project: Flink Issue Type: Sub-

[jira] [Created] (FLINK-1982) Remove dependencies of Record for Flink runtime and core

2015-05-06 Thread Henry Saputra (JIRA)
Henry Saputra created FLINK-1982: Summary: Remove dependencies of Record for Flink runtime and core Key: FLINK-1982 URL: https://issues.apache.org/jira/browse/FLINK-1982 Project: Flink Issue

[jira] [Created] (FLINK-1981) Add GZip support

2015-05-06 Thread Sebastian Kruse (JIRA)
Sebastian Kruse created FLINK-1981: -- Summary: Add GZip support Key: FLINK-1981 URL: https://issues.apache.org/jira/browse/FLINK-1981 Project: Flink Issue Type: New Feature Componen

[jira] [Created] (FLINK-1980) Allowing users to decorate input streams

2015-05-06 Thread Sebastian Kruse (JIRA)
Sebastian Kruse created FLINK-1980: -- Summary: Allowing users to decorate input streams Key: FLINK-1980 URL: https://issues.apache.org/jira/browse/FLINK-1980 Project: Flink Issue Type: New Fe

[jira] [Created] (FLINK-1979) Implement Loss Functions

2015-05-06 Thread JIRA
Johannes Günther created FLINK-1979: --- Summary: Implement Loss Functions Key: FLINK-1979 URL: https://issues.apache.org/jira/browse/FLINK-1979 Project: Flink Issue Type: Improvement

RE: NioEventLoop consumes most of the CPU

2015-05-06 Thread Kruse, Sebastian
Hi, Thanks for your answers. I see the point that this might be an issue with the profiler, although I cannot see the epoll function in the stack trace. It seems that the NioEventLoop is not using this function but rather seems to work around it (cf. NioEventLoop:220). However, are you using an

Re: NioEventLoop consumes most of the CPU

2015-05-06 Thread Stephan Ewen
Uful has a good point The NIO epoll wait method leaves the thread in state RUNNABLE. That may explain things. Still, would be good to have more information on your setup. Stephan Am 06.05.2015 10:15 schrieb "Ufuk Celebi" : > I agree with Stephan's points. Thanks for reporting and let's investig

Re: NioEventLoop consumes most of the CPU

2015-05-06 Thread Ufuk Celebi
I agree with Stephan's points. Thanks for reporting and let's investigate this further. To keep in mind: I think VisualVM is using hprof for CPU sampling, which has some known issues ( http://www.brendangregg.com/blog/2014-06-09/java-cpu-sampling-using-hprof.html). For one thing, it's profiling Ja

[jira] [Created] (FLINK-1978) POJO serialization NPE

2015-05-06 Thread Flavio Pompermaier (JIRA)
Flavio Pompermaier created FLINK-1978: - Summary: POJO serialization NPE Key: FLINK-1978 URL: https://issues.apache.org/jira/browse/FLINK-1978 Project: Flink Issue Type: Bug Comp

[jira] [Created] (FLINK-1977) Rework Stream Operators to always be push based

2015-05-06 Thread Aljoscha Krettek (JIRA)
Aljoscha Krettek created FLINK-1977: --- Summary: Rework Stream Operators to always be push based Key: FLINK-1977 URL: https://issues.apache.org/jira/browse/FLINK-1977 Project: Flink Issue Typ