Hi Ismael, Thanks for the feedback. I noticed that the graalvm-reachability-metadata project now includes > kafka-clients[1]. Worth checking if the metadata over there matches what > we've observed for our own attempt. > After reviewing the graalvm-reachability-metadata, I observed that they have provided configurations solely for the kafka-clients. However, we require configurations for the server (the core module). To obtain these configurations, I've leveraged the existing Apache Kafka system tests. I'm executing the tests using GraalVM JIT with the Native-Image Agent attached. I merge the configurations once all the tests have been completed.
Can we please add peak memory usage as one of the columns of the table? > I see that the information is covered elsewhere - it would be nice to have > it as part of the table too. > Done GraalVM for JDK 21 was recently released, can we use that? Does it improve > the numbers in a meaningful way? Sure, we will be using the Image corresponding to 21. The community version Image: https://github.com/graalvm/container/pkgs/container/graalvm-community/129309492?tag=21 It would be good to include details on the naming - how will this image > be named to differentiate itself from the non graalvm one? > Added a section in the respective KIPs. In summary, we propose naming the image as *<image-name>:<kafka-version>-<optional-suffix>*. For the native AK docker image, we are considering '*kafka-local*' as it clearly signifies that this image is intended exclusively for local development purposes. Another option under consideration is '*kafka-native* '. Can we talk a bit more about the compatibility guarantees while this > image is still experimental? Do you here mean the functionalities that this native AK docker image will be able to support? I have updated the KIP with the above points. Regards, Krishna On Wed, Sep 20, 2023 at 11:39 PM Ismael Juma <m...@ismaeljuma.com> wrote: > Hi Krishna, > > Thanks for the KIP. The results are impressive! A few quick comments: > > 1. I noticed that the graalvm-reachability-metadata project now includes > kafka-clients[1]. Worth checking if the metadata over there matches what > we've observed for our own attempt. > 2. Can we please add peak memory usage as one of the columns of the table? > I see that the information is covered elsewhere - it would be nice to have > it as part of the table too. > 3. GraalVM for JDK 21 was recently released, can we use that? Does it > improve the numbers in a meaningful way? > 4. Can we talk a bit more about the compatibility guarantees while this > image is still experimental? > 5. It would be good to include details on the naming - how will this image > be named to differentiate itself from the non graalvm one? > > [1] > > https://github.com/oracle/graalvm-reachability-metadata/tree/df53148e4b2f730ad692daa956d72ab0cd7d43e3/metadata/org.apache.kafka/kafka-clients/3.5.1 > > Ismael > > On Fri, Sep 8, 2023 at 12:17 AM Krishna Agarwal < > krishna0608agar...@gmail.com> wrote: > > > Hi, > > I want to submit a KIP to deliver an experimental Apache Kafka docker > > image. > > The proposed docker image can launch brokers with sub-second startup time > > and minimal memory footprint by leveraging a GraalVM based native Kafka > > binary. > > > > KIP-974: Docker Image for GraalVM based Native Kafka Broker > > < > > > https://cwiki.apache.org/confluence/display/KAFKA/KIP-974%3A+Docker+Image+for+GraalVM+based+Native+Kafka+Broker > > > > > > > Regards, > > Krishna > > >
