High CPU utilization due to massive other IOPS
Applies to
- ONTAP 9
- NFS
Issue
- CPU utilization higher than CPU peak performance line in the node.
- CPU utilization is near 100%.
- No obvious high write/read latency shown on volume from CPU D-blade.
Example: Node 18 has a high CPU utilization due to user workload, but other nodes of the cluster are idle/barely utilized as seen in the node shell sysstat -x 1
command.
Note: Columns removed to improve readability
Cluster::> node run node18 sysstat -x 1 CPU NFS CIFS HTTP Total Net kB/s Disk kB/s in out read write 100% 65782 0 0 65782 14918 2953948 1325731 14920 99% 64522 0 0 64522 13125 2743195 1256472 12126 99% 64656 0 0 64656 12121 2735679 1211287 12120 100% 65898 0 0 65989 14812 2948834 1389425 14815 Cluster::> qos statistics volume latency show Workload ID Latency Network Cluster Data Disk QoS NVRAM --------------- ------ ---------- ---------- ---------- ---------- --------- --------- ---------- total- - 4.49ms 99.00us 70.00us 4.27ms 153.00us 0ms 0ms FG_vol18_001.. 4201 4.05ms 90.00us 0ms 4.046ms 150.00us 0ms 0ms