Re: Subscription to mailing list

Hi, You can subscribe to the mailing list by sending a email to dev-subscr...@flink.apache.org . Other mailing lists about Flink are in http://flink.apache.org/community.html#mailing-lists . Regards. C

Re: Subscription to mailing list

Hi, You can subscribe to the mailing list by sending a email to dev-subscr...@flink.apache.org . Other mailing lists about Flink are in http://flink.apache.org/community.html#mailing-lists Regards. Ch

Subscription to mailing list

Hi, I would like to subscribe to the Apache Flink developer mailing list. Regards, Devesh Gade. -- Tough times dont last,Tough People Do.

Re: [Delta Iterations] The dirty insides(insights)

You can always have the vertex value being a tuple3 and let the user set only a part of it (the value). It would work the same way as with the vertex key - the user function can see it, but not change it. On Sat, Mar 21, 2015 at 8:42 PM, Andra Lungu wrote: > Hi Stephan, > > You got the idea! I

Re: [GSoc][flink-streaming] Interested in pursuing FLINK-1617 and FLINK-1534

Thanks for the explanation Marton. I've decided to try out for FLINK-1534. After reading through the thesis[4] and a few other papers[1][2][3], I believe I've gathered a little context to ask more questions. But I'm still not sure how Flink's internals work so please bear with me. Although the ong

Re: [Delta Iterations] The dirty insides(insights)

Hi Stephan, You got the idea! I wanted to make the inDeg and outDeg part of the value itself(that was one of my first versions), but then I thought it would be too annoying for a user who does not want to access the degrees in vertexUpdate/ sendMessages to carry them around. Because in this scena

Re: [Delta Iterations] The dirty insides(insights)

Hi Andra! I am still not 100% sure about the cause of the problem, but here is a bit of background on how we model the computation as a delta iteration and how you can extend the vertex state. Maybe that helps answer your question. When using delta iterations for vertex-centric iterations, we kee

Re: Queries regarding RDFs with Flink

Hi Flavio! I see initially two ways of doing this: 1) Do a series of joins. You start with your subject and join two or three times using the "objects-from-triplets == subject" to make one hop. You can filter the verbs from the triplets before if you are only interested in a special relationship.

Re: Restructuring the maven projects

Optimizer and Compiler can stay separate, of course. I was just thinking that we have a lot of projects and this would be a simple way to get rid of one. But it actually is intuitive to me as well to keep them separate. On Wed, Mar 18, 2015 at 1:23 AM, Fabian Hueske wrote: > I agree that it's a

Re: Improve the documentation of the Flink Architecture and internals

Very nice post, Till! We are starting to get much better with this... On Sat, Mar 21, 2015 at 6:45 PM, Henry Saputra wrote: > Awesome, thanks Till > > On Saturday, March 21, 2015, Till Rohrmann wrote: > > > I wrote some internal documentation for Akka and the distributed > > communication [1].

Re: Improve the documentation of the Flink Architecture and internals

Awesome, thanks Till On Saturday, March 21, 2015, Till Rohrmann wrote: > I wrote some internal documentation for Akka and the distributed > communication [1]. > > Cheers, > > Till > > [1] https://cwiki.apache.org/confluence/display/FLINK/Akka+and+Actors > > On Fri, Mar 20, 2015 at 7:31 PM, Henry

Re: [DISCUSS] Name of Expression API and DataSet abstraction

Both names look ok but if I had to choose I would go for the Relation API. It defines a new DSL in contrast to data types (ie DataSet, DataStream) so it doesn’t have to follow the same convention. Plus, it is a single word with sufficient meaning. Paris On 21 Mar 2015, at 17:29, Fabian Hueske m

Re: [DISCUSS] Name of Expression API and DataSet abstraction

I'm in favor of Relation API because the API operates on a set of tuples with a fixed (flat) schema which is known as relation from relational databases. Everybody who took a database intro class should be familiar with this term. On Mar 21, 2015 5:14 PM, "Henry Saputra" wrote: > I like the Rela

Re: [DISCUSS] Name of Expression API and DataSet abstraction

OK, as recommended by Fabian, I explain why I like Relational better =) The concept of Relational suggest the concept of querying data based on the relationship of different sources and the predicates/ filters attached to them. Or we could go more like keyword name such as "REMOQ" which stands fo

Re: Improve the documentation of the Flink Architecture and internals

I wrote some internal documentation for Akka and the distributed communication [1]. Cheers, Till [1] https://cwiki.apache.org/confluence/display/FLINK/Akka+and+Actors On Fri, Mar 20, 2015 at 7:31 PM, Henry Saputra wrote: > Ah the Tweet infra bot just announce extended downtime for Confluence

Re: [DISCUSS] Name of Expression API and DataSet abstraction

I like the Relation or Relational. So maybe we could use DataRelation as the abstraction? - Henry On Mon, Mar 16, 2015 at 9:30 AM, Fabian Hueske wrote: > I am also more in favor of Rel and Relation, but DataTable nicely follows > the terms DataSet and DataStream. > On Mar 16, 2015 4:58 PM, "Alj

[jira] [Created] (FLINK-1766) Fix the bug of equals function of FSKey

Sibao Hong created FLINK-1766: - Summary: Fix the bug of equals function of FSKey Key: FLINK-1766 URL: https://issues.apache.org/jira/browse/FLINK-1766 Project: Flink Issue Type: Bug R