Comparison with trino + iceberg?

[JakkuSakura]

I found databend and trino + iceberg pretty much do the DWH.

Pros and cons for them?

[BohuTANG]

@Xuanwo is working on iceberg-rust and integrating it into Databend to enhance its lakehouse capabilities. With Databend's comprehensive features, you might not even need Trino—it has everything you need! Could you share more about your requirements?

[JakkuSakura]

I'm exploring solutions for quantitative finance, especially in the research area. We need to save and query a large amount of data without affecting performance.
After work on iceberg-rust, databend and trino should all work perfectly on iceberg. Then the comparison becomes the execution layer(i.e. quality of push-down) and ecosystem. I can try out both then

[JakkuSakura]

Indeed, databend has everything I need. I tried to setup trino + iceberg, at first I thought I need trino + iceberg + S3(AWS S3/seaweedfs/minio), but I also need a metadata service(hive + hadoop/AWS glue), which are too much for individuals or small teams. I decided to use databend, and wait for the iceberg writing feature