Hi folks, we are evaluating removing file prefetching (https://github.com/apache/datafusion/pull/20916) because:
1. It seems to have a negative impact even against high latency data stores. 2. It adds complexity that gets in the way of future developments (namely morselized scans which show huge perf improvements). Does anyone knowingly rely on this prefetch behavior in any way? --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
