Use this procedure to determine if you need to add more disks. Note that disks can be over-utilized by making a poor choice for the RAID profile and/or ZFS record size. In this case, it may be possible to reduce existing disk utilization by moving from RAIDZ to mirrored profiles and/or matching ZFS record sizes to client I/O sizes.
In addition, if the system is configured with no read- or write-optimized flash drives, then all of the I/O operations beyond DRAM are served by disks. For better performance, consider using flash for any workload that includes random reads or synchronous writes.
hostname:analytics worksheets> select worksheet-000 hostname:analytics worksheet-000> dataset
hostname:analytics worksheet-000 dataset (uncommitted)> set name="io.disks[utilization=70]" name = io.disks[utilization=70] hostname:analytics worksheet-000 dataset (uncommitted)> commit
hostname:analytics worksheet-000> done hostname:analytics worksheets> done
hostname:> analytics datasets
hostname:analytics datasets> show Datasets: DATASET STATE INCORE ONDISK NAME dataset-000 active 1.27M 15.5M arc.accesses[hit/miss] dataset-001 active 517K 9.21M arc.accesses[hit/miss=metadata hits][L2ARC eligibility] ... dataset-025 active 290K 7.80M io.disks[utilization=70] hostname:analytics datasets>
In this example, dataset name io.disks[utilization=70] corresponds to dataset-025.
hostname:analytics datasets> select dataset-025
hostname:analytics dataset-025> read 1800
You may want to use more disks when at least 50% of existing disks are at least 70% utilized.