You can always use as much disk resource as available, the blk-weight only kicks in if the disk is over-provisioned to manage sharing between different instances on the same allocator
It uses the same weighting as the CPU (ie changing one cluster's CPU weight will also increase its IO weight as discussed here)
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.