mapleFU commented on issue #38275: URL: https://github.com/apache/arrow/issues/38275#issuecomment-1763453732
What's the version of arrow are you using? 1. During read a single ipc file, there is a `use_thread` argument, which enable using the user-passed executor or system default executor to decode the batch. It's default enabled in current master code. 2. If you only want some prefetch in a large set of files, you can try to put `pa.ipc.open_stream(memory_mapped_stream).read_all()` or other to the thread pool, rather than doing in the current way. 3. You can also try to use `dataset` api ( https://arrow.apache.org/docs/python/dataset.html#reading-datasets ), it enable some prefetch here. -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
