Hi xubo, +1 Thank you for suggesting this. Totally agree with it. Containerising carbon data is much needed one.
Few inputs to consider 1. Since most of the users are not using spark 2.4 and spark 2.4 we have to put effort in removing the support. 2. We can support docker container for spark 3 only. Regards Akash On Mon, 10 Apr 2023 at 12:13 AM, Bo Xu <xubo...@apache.org> wrote: > Using docker to run code is a very popular development method for users. > Apache spark docker image has more then 1000000 downloads, > jupyter/all-spark-notebook docker images has more then 50000000 downloads > > > On 2023/04/09 18:35:44 徐波 wrote: > > For newcomers of CarbonData, it's not very easy to build CarbonData > > environment and run CarbonData. > > So we should improve better usability for CarbonData. > > There are some directions, we can discuss: > > 1. Provide docker image of CarbonData, we can run carbondata very easily. > > refer to spark: https://hub.docker.com/r/apache/spark > > 2.provide notebook docker image with CarbonData, we can run carbondata > > very easily for interactive scenario. refer to spark: > > https://hub.docker.com/r/jupyter/all-spark-notebook > > ... > > > > Please discuss if you have better idea > > >
