Skip to main content
A newer release of this product is available.

qos statistics workload latency show

Contributors
Suggest changes

Display latency breakdown data per QoS workload

Availability: This command is available to cluster administrators at the admin privilege level.

Description

The qos statistics workload latency show command displays the average latencies for QoS workloads on Data ONTAP subsystems.

The command displays the following data:

  • The QoS workload name (Workload)

  • The QoS workload ID (ID)

  • Total latency observed per I/O operation (Latency)

  • Latency observed per I/O operation in the Network subsystem (Network)

  • Latency observed per I/O operation across the internally connected nodes in a Cluster (Cluster)

  • Latency observed per I/O operation in the Data management subsystem (Data)

  • Latency observed per I/O operation in the Storage subsystem (Disk)

  • Latency observed per I/O operation for QoS Policy Group Ceiling (QoS Max)

  • Latency observed per I/O operation for QoS Policy Group Floor (QoS Min)

  • Latency observed per I/O operation for NVRAM transfer (NVRAM)

  • Latency observed per I/O operation for Object Store(Cloud) operations

  • Latency observed per I/O operation for FlexCache (FlexCache) operations

  • Latency observed per I/O operation for Synchronous Snapmirror (SM Sync) operations

  • Latency observed per I/O operation for Volume Activation (VA) operations

The results displayed per iteration are sorted by the total latency field. Each iteration starts with a row that displays the average latency, in microseconds (us) or milliseconds (ms) observed across all QoS workloads.

Parameters

[-node {<nodename>|local}] - Node

Selects the QOS workloads that match this parameter value. If you do not specify this parameter, the command displays data for the entire cluster.

[-iterations <integer>] - Number of Iterations

Specifies the number of times that the command refreshes the display with updated data before terminating. If you do not specify this parameter, the command continues to run until you interrupt it by pressing Ctrl-C.

[-refresh-display {true|false}] - Toggle Screen Refresh Between Each Iteration

Specifies the display style. If true, the command clears the display after each data iteration. If false, the command displays each data iteration below the previous one. The default is false.

{ [-rows <integer>] - Number of Rows in the Output

Specifies the number of busiest QoS policy groups to display. Valid values are from 1 to 20. The default value is 10.

[-policy-group <text>] - QoS Policy Group Name

Selects the QoS workloads that belong to the QoS policy group specified by this parameter value. If you do not specify this parameter, the command displays data for all QoS workloads.

| [-adaptive-policy-group <text>] - Adaptive QoS Policy Group Name

Selects the QoS workloads that belong to the QoS adaptive policy group specified by this parameter value. If you do not specify this parameter, the command displays data for all QoS workloads.

| [-workload <text>] - QoS Workload Name

Selects the QoS workload that match this parameter value. If you do not specify this parameter, the command displays data for all QoS workloads.

| [-workload-id <integer>] - QoS Workload ID }

Selects the QoS workload that match the QoS workload ID specified by this parameter value.

[-show-flexgroup-as-constituents {true|false}] - Display Flexgroups as Constituents

If the parameter is specified and if the value is true, it will display data for FlexVols and Flexgroup Constituents. Otherwise it will display data for FlexVols and Flexgroups.

Examples

cluster1::> qos statistics workload latency show -iterations 100 -rows 3
Workload            ID  Latency    Network  Cluster       Data     Disk  Qos Max    Qos Min      NVRAM      Cloud  FlexCache    SM Sync         VA
--------------- ------ --------   -------- --------   -------- -------- -------- ---------- ---------- ---------- ---------- ---------- ----------
-total-                110.35ms   110.02ms      0ms   327.00us      0ms      0ms        0ms        0ms        0ms        0ms        0ms        0ms
vs1vol0            111 167.82ms   167.22ms      0ms   603.00us      0ms      0ms        0ms        0ms        0ms        0ms        0ms        0ms
vol1              1234 117.76ms   117.56ms      0ms   191.00us      0ms      0ms        0ms        0ms        0ms        0ms        0ms        0ms
vol2               999  44.24ms    44.05ms      0ms   190.00us      0ms      0ms        0ms        0ms        0ms        0ms        0ms        0ms
-total-             -   38.89ms    38.63ms      0ms   256.00us      0ms      0ms        0ms        0ms        0ms        0ms        0ms        0ms
vol2               999  64.47ms    64.20ms      0ms   266.00us      0ms      0ms        0ms        0ms        0ms        0ms        0ms        0ms
vol1              1234  27.28ms    27.03ms      0ms   253.00us      0ms      0ms        0ms        0ms        0ms        0ms        0ms        0ms
vs1vol0            111  23.72ms    23.47ms      0ms   249.00us      0ms      0ms        0ms        0ms        0ms        0ms        0ms        0ms
-total-              - 409.81ms   409.65ms      0ms   169.00us      0ms      0ms        0ms        0ms        0ms        0ms        0ms        0ms
vol1              1234 816.92ms   816.80ms      0ms   120.00us      0ms      0ms        0ms        0ms        0ms        0ms        0ms        0ms
vol2               999 407.88ms   407.66ms      0ms   219.00us      0ms      0ms        0ms        0ms        0ms        0ms        0ms        0ms
vs1vol0            111   3.68ms     3.49ms      0ms   193.00us      0ms      0ms        0ms        0ms        0ms        0ms        0ms        0ms

The example above displays latencies for the 3 QoS workloads with the highest latencies and it refreshes the display 100 times before terminating.

cluster1::> qos statistics workload latency show -iterations 100 -rows 2 -policy-group pg1
Workload            ID    Latency    Network    Cluster       Data       Disk    Qos Max    Qos Min      NVRAM      Cloud  FlexCache    SM Sync         VA
--------------- ------ ---------- ---------- ---------- ---------- ---------- ---------- ---------- ---------- ---------- ---------- ---------- ----------
-total-              -     4.80ms   287.00us        0ms   427.00us     4.08ms        0ms        0ms        0ms        0ms        0ms        0ms        0ms
file-test1-wi..   7872     9.60ms   265.00us        0ms   479.00us     8.85ms        0ms        0ms        0ms        0ms        0ms        0ms        0ms
file-test1_a-..   6437     8.22ms   262.00us        0ms   424.00us     7.53ms        0ms        0ms        0ms        0ms        0ms        0ms        0ms
-total-              -     4.20ms   296.00us        0ms   421.00us     3.48ms        0ms        0ms        0ms        0ms        0ms        0ms        0ms
file-test1-wi..   7872     8.70ms   211.00us        0ms   489.00us     8.00ms        0ms        0ms        0ms        0ms        0ms        0ms        0ms
file-test1_a-..   6437     6.70ms   297.00us        0ms   464.00us     5.94ms        0ms        0ms        0ms        0ms        0ms        0ms        0ms
-total-              -     5.90ms   303.00us        0ms     1.71ms     3.88ms        0ms        0ms        0ms        0ms        0ms        0ms        0ms
file-test1-wi..   7872    11.36ms   263.00us        0ms     2.06ms     9.04ms        0ms        0ms        0ms        0ms        0ms        0ms        0ms
file-test1_a-..   6437     9.48ms   250.00us        0ms     2.30ms     6.93ms        0ms        0ms        0ms        0ms        0ms        0ms        0ms

The example above displays latencies for the 2 QoS workloads belonging to QoS policy group pg1 with the highest IOPS and it refreshes the display 100 times before terminating.

cluster1::> qos statistics workload latency show -iterations 100 -workload-id 9492
   Workload            ID    Latency    Network    Cluster       Data       Disk    Qos Max    Qos Min      NVRAM      Cloud  FlexCache    SM Sync         VA
   --------------- ------ ---------- ---------- ---------- ---------- ---------- ---------- ---------- ---------- ---------- ---------- ---------- ----------
   -total-              -   443.00us   273.00us        0ms   170.00us        0ms        0ms        0ms        0ms        0ms        0ms        0ms        0ms
   file-test1_b-..   9492   440.00us   272.00us        0ms   168.00us        0ms        0ms        0ms        0ms        0ms        0ms        0ms        0ms
   -total-              -   577.00us   313.00us        0ms   264.00us        0ms        0ms        0ms        0ms        0ms        0ms        0ms        0ms
   file-test1_b-..   9492   607.00us   316.00us        0ms   291.00us        0ms        0ms        0ms        0ms        0ms        0ms        0ms        0ms
   -total-              -   475.00us   291.00us        0ms   184.00us        0ms        0ms        0ms        0ms        0ms        0ms        0ms        0ms
   file-test1_b-..   9492   476.00us   293.00us        0ms   183.00us        0ms        0ms        0ms        0ms        0ms        0ms        0ms        0ms
   -total-              -   628.00us   284.00us        0ms   344.00us        0ms        0ms        0ms        0ms        0ms        0ms        0ms        0ms
   file-test1_b-..   9492   591.00us   281.00us        0ms   310.00us        0ms        0ms        0ms        0ms        0ms        0ms        0ms        0ms

The example above displays the latencies for the QoS workload with QoS workload ID 9492 and it refreshes the display 100 times before terminating.