@mperham @bpot @mxavier also @nathanmarz Storm shows relying on ZK in this context isn't a bottleneck
-
-
@bascule@mperham@mxavier@nathanmarz kafka is planning to move away from zk for offset updates for perf reasons https://issues.apache.org/jira/browse/KAFKA-657 … -
@bpot@mperham@mxavier@nathanmarz it's storm-kafka that does the offset tracking -
@bascule@bpot@mperham@nathanmarz this all sounds very straightforward to implement and deploy for simple job processing. Thanks. -
@mxavier@bpot@mperham@nathanmarz yep, I implemented as much at@livingsocial but never open sourced it. Should reimplement ;)
End of conversation
New conversation -
-
-
@bascule@mperham@mxavier@nathanmarz I think storm and trident usually operate on batches of messages not individual ones -
@bpot@mperham@mxavier@nathanmarz yes and this is a good thing
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.