kou commented on issue #38275: URL: https://github.com/apache/arrow/issues/38275#issuecomment-1772154032
> When user want to loading all data to memory-mapped RecordBatches, how could this benefit the reading time? Does it mean that is the stream format useful with mmap, right? If a user reads all record batches, process them and drops them, it will not be useful. In the "reads all record batches" phase, some record batches may be page outed. In the "process them" phase, the page outed record batched will be re-mmapped. If a user reads a record batch, processes the record batch and drops the record batch, it will be useful. A mmapped record batch in the "reads a record batch" phase will not be page outed in the "process the record batch" phase. -- 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]
