+1 concise and intuitive way to define a DAG. Of course we will also need to build out the underlying operators. What about iteration, join or window based operations?
On Mon, Feb 29, 2016 at 9:05 PM, Bhupesh Chawda <[email protected]> wrote: > +1 for transforming Java 8 streams to an Apex Dag. > This can act as an implementation of Java 8 streams for Big Data. > > -Bhupesh > > > On Tue, Mar 1, 2016 at 8:18 AM, David Yan <[email protected]> wrote: > > > I'm not a Java 8 expert, but it would be extremely valuable if we can > > support the Java 8 stream API with by providing an implementation to > > transform Java 8 streams to Apex topology. > > > > On Mon, Feb 29, 2016 at 3:30 PM, Siyuan Hua <[email protected]> > > wrote: > > > > > Hi community, > > > > > > Apex, Spark Streaming and Flink, every stream processing framework has > > it's > > > own API to build a DAG. They all have pros and cons, but we also see > many > > > similar concepts and APIs between Flink & Sparking Streaming. We want > to > > > have this kind of similar APIs so that other developers can easily > learn > > > and convert other apps to Apex. > > > > > > Here is the link to the proposal > > > > > > > > > https://docs.google.com/document/d/163LmQjX860b61NDe3ZzR0hRTPtE-4GF0iHaVhmHQssY/edit#heading=h.aytn6rz7u1e4 > > > Here is the ticket > > > https://issues.apache.org/jira/browse/APEXMALHAR-2006 > > > > > > Any suggestions are welcome. > > > > > > > > > Best, > > > Siyuan > > > > > >
