nevi-me commented on pull request #9329: URL: https://github.com/apache/arrow/pull/9329#issuecomment-802247250
> does anyone knows why we are using Arc<ArrayData> in all arrays? I think it comes down just to memory size, as cloning `Arc<T>` would give us a pointer-sized variable, compared to cloning `T` including its `DataType`. > Do you envision an issue with removing the Arc? I personally don't. The underlying bytes are still backed by `Arc`, so we shouldn't use a lot more memory for copies of arrays. The perf benchmarks are also positive. > Would someone be so kind and run the benches independently, just to be sure. I attached my set of benchmark results, but I think the datafusion benchmarks would be more meaningful. @Dandandan would you be able to run datafusion benchmarks from this branch vs master, and share results? Thanks -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org
