I don’t think “benchmarking” frameworks WRT Kafka is a particularly
informative. The various frameworks available are better compared WRT their
features and processing limitations. For example, Akka-streams for kafka
effects a more intuitive way to express asynchronous operations. If you were
to benchmark each framework with a simple poll-transformation-publish workload,
I think you would find very little difference between them (assuming that they
were all configured appropriately…minimum consumer bytes setting for instance).
I think each framework would be better evaluated according to it’s
features….just my thoughts.
-David
On 3/23/17, 9:38 AM, "Eno Thereska" <eno.there...@gmail.com> wrote:
Hi Giselle,
Great idea! In Kafka Streams we have a few micro-benchmarks we run nightly.
They are at:
https://github.com/apache/kafka/blob/trunk/streams/src/test/java/org/apache/kafka/streams/perf/SimpleBenchmark.java
<https://github.com/apache/kafka/blob/trunk/streams/src/test/java/org/apache/kafka/streams/perf/SimpleBenchmark.java>
It's mostly simple stuff (aggregations, joins) and we are continuously
updating them and adding more.
The nightly results are kept publicly at
http://testing.confluent.io/confluent-kafka-system-test-results/
<http://testing.confluent.io/confluent-kafka-system-test-results/>, e.g., see
report on 2017-03-21:
http://confluent-kafka-system-test-results.s3-us-west-2.amazonaws.com/2017-03-21--001.1490119830--apache--trunk--05690f0/report.html
<http://confluent-kafka-system-test-results.s3-us-west-2.amazonaws.com/2017-03-21--001.1490119830--apache--trunk--05690f0/report.html>
(search for "simple_benchmark_test").
Your feedback and input is always appreciated.
Thanks,
Eno
> On 23 Mar 2017, at 10:09, Giselle van Dongen <giselle.vandon...@ugent.be>
wrote:
>
> Dear users of Streaming Technologies,
>
> As a PhD student in big data analytics, I am currently in the process of
> compiling a list of benchmarks (to test multiple streaming frameworks) in
> order to create an expanded benchmarking suite. The benchmark suite is
being
> developed as a part of my current work at Ghent University.
>
> The included frameworks at this time are, in no particular order, Spark,
> Flink, Kafka (Streams), Storm (Trident) and Drizzle. Any pointers to
> previous work or relevant benchmarks would be appreciated.
>
> Best regards,
> Giselle van Dongen
