sapant-msft opened a new pull request #1708: HADOOP-16696: Always read ahead config, to use read ahead even for non sequential reads. URL: https://github.com/apache/hadoop/pull/1708 Adding a config alwaysReadAhead, set to False by default, to be able to use abfs' read ahead capability, even for non-sequential reads. At the moment, only sequential reads support read ahead. A read ahead is queued only after a sequential read is made, so we miss out on gains where we have a non-sequential followed by a sequential read. For example, seek(n), read 1 byte, read 10 bytes , seek (m), could benefit from read ahead, but not currently supported.
---------------------------------------------------------------- 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: [email protected] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
