Protocols: Average Latency Statistics
Average latency statistics are available for the Fibre Channel, iSCSI, NFS, SMB, and SRP protocols:
Average latency statistics provide average latency per second. A pair of timestamp values is used to determine the time of the operation, and then add the elapsed time to the average aggregation. Use these statistics to characterize performance issues associated with these protocols. Average latency can be broken down by type of operation and pool.
When to Check Average Latency
Monitor average latency if you suspect performance issues. For example, you might want to generate an alert if the average latency substantially increases for a period of time. Average latency can be viewed on the dashboard.
Further Analysis
See also “Protocols” in Default Statistics.
Protocol: Fibre Channel Average Latency
Table 6-34 Breakdowns of Fibre Channel Average Latency
Breakdown | Description |
---|---|
type of operation |
Fibre Channel operation performed, such as read or write. |
Further Analysis
-
Protocol: Fibre Channel Bytes for FC bytes/second
-
Protocol: Fibre Channel Operations for numerous other breakdowns on FC operations
-
Cache: ARC Accesses to learn how well an FC read workload is returning from cache
-
Disk: I/O Operations for the back-end disk I/O caused
Protocol: iSCSI Average Latency
Table 6-35 Breakdowns of iSCSI Average Latency
Breakdown | Description |
---|---|
type of operation |
iSCSI operation performed, such as read or write. |
pool |
Storage pool on which the operation is performed. |
Further Analysis
-
Protocol: iSCSI Bytes for iSCSI bytes/second
-
Protocol: iSCSI Operations for numerous other breakdowns on iSCSI operations
-
Cache: ARC Accesses to learn how well an iSCSI read workload is returning from cache
-
Disk: I/O Operations for the back-end disk I/O caused
Protocol: NFSv[2-4] Average Latency
Supported NFS versions are: NFSv2, NFSv3, NFSv4.0, and NFSv4.1.
Table 6-36 Breakdowns of NFS Average Latency
Breakdown | Description |
---|---|
type of operation |
NFS operation performed, such as read or write. |
pool |
Storage pool on which the NFS operation is performed. |
Further Analysis
-
Protocol: NFSv[2-4] Bytes for NFSv[2-4] bytes/second
-
Protocol: NFSv[2-4] Operations for numerous other breakdowns on NFS operations
-
Network: Device Bytes for a measure of network throughput caused by the NFS activity
-
Cache: ARC Accesses to learn how well an NFS read workload is returning from cache
-
Disk: I/O Operations for the back-end disk I/O caused
Protocol: SMBv[1-3] Average Latency
Supported SMB versions are SMB, SMB2, and SMB3.
Table 6-37 Breakdowns of SMB Average Latency
Breakdown | Description |
---|---|
type of operation |
SMB operation performed, such as read or write. |
pool |
Storage pool on which the SMB operation is performed. |
Further Analysis
-
Protocol: SMBv[1-2] Bytes for SMB bytes/second
-
Protocol: SMB Operations for numerous other breakdowns on SMB operations
-
Cache: ARC Accesses to learn how well an SMB read workload is returning from cache
-
Disk: I/O Operations for the back-end disk I/O caused
Protocol: SRP Average Latency
Table 6-38 Breakdowns of SRP Latency
Breakdown | Description |
---|---|
type of operation |
SRP operation performed, such as read or write. |
Further Analysis
-
Protocol: SRP Bytes for SRP bytes/second
-
Protocol: SRP Operations for numerous other breakdowns on SRP operations
-
Cache: ARC Accesses to learn how well an SRP read workload is returning from cache
-
Disk: I/O Operations for the back-end disk I/O caused