Howie Wang

If this feature is to prevent duplicate records, it is not very helpful, because we can add a view on top of BQ table with deduplication logic. If this feature...

Agree with @ttyusupov , in commit https://github.com/apache/avro/commit/7e4037de2a891fa738aaf8a4fc56f424d6c6833a#diff-c7934590c625ba67bf1b2ad5511a4f58L167, a previous statement that aim to set default value was removed. However, I don't see it was mentioned anywhere in its ticket, https://issues.apache.org/jira/browse/AVRO-388....

Hi @ewencp , thanks for the feedback! I am excited to see the potential use cases. As @criccomini said, it will be great if you can provide some semi-detailed implementation...

Waiting for the fix for `@cluster`, https://github.com/confluentinc/ducktape/pull/207

@ewencp Thank you so much for your comment. I try to make a brief summary for each sections and provide some of my thought, as below: **Q1**: You should just...

Hi @jarekr would you please leave a comment! 😃Thanks! cc @criccomini

> A1: Yeah, that's roughly what I meant. I think we should consider dropping the need for using a ClusterSpec method -- can we look into whether it's feasible to...

> So I had been thinking more along the lines of constraints that cluster resource managers tend to enforce, e.g. (cpu >= 2vcpu, mem >= 2GB), etc. Sounds good to...

@ewencp Supported node allocation based on (multi-dimension) machine-type according to previous discussion. Before we fixing all the tests, just want to get some feedbacks from you. ``` # Vagrantfile defines...

@ewencp Based on comment, I propose a more user-friendly pattern. ``` STORAGE_NODE = NodeSpec(LINUX, MachineType(cpu=1, disk=100G)) cluster_spec = ClusterSpec.from_nodes(STORAGE_NODE.with_num_nodes(3)) s = StorageService(cluster_spec) ``` ``` @cluster(cluster_spec=ClusterSpec.from_nodes(STORAGE_NODE.with_num_nodes(3)), SERVER_NODE.with_num_nodes(2)) def my_test() """ A...