Table of ContentsView in Frames

Controlling and monitoring workload performance

You control and monitor workload performance to address performance problems and to proactively limit workloads that have defined performance targets.

Before you begin

Steps

  1. Identify the storage objects that you want to assign to Storage QoS policy groups.
    A best practice is to assign the same type of storage object to all policy groups.
  2. Use the qos policy-group create command to create a new policy group or use the qos policy-group modify command to modify an existing policy group.
    You can specify a maximum throughput limit when you create the policy group or you can wait until after you monitor the workload. Monitoring the workload first can help you identify the limit that you need to set. If you do not specify a maximum throughput, the workloads get best-effort performance.
    Example
    The following command creates policy group pg-vs1 with a maximum throughput of 5,000 IOPS.
    cluster1::> qos policy-group create pg-vs1 -vserver vs1 -max-throughput 5000iops
    Example
    The following command creates policy group pg-app2 without a maximum throughput.
    cluster1::> qos policy-group create pg-app2 -vserver vs2
  3. To assign a storage object to a policy group, use the create or modify command for the SVM, volume, LUN, or file.
    Example
    The following command assigns the SVM vs1 to policy group pg-vs1.
    cluster1::> vserver modify -vserver vs1 -qos-policy-group pg-vs1
    Example
    The following command creates the volume app2 and assigns it to policy group pg-app2.
    cluster1::> volume create -vserver vs2 -volume app2 -aggregate aggr2 -qos-policy-group pg-app2
  4. To identify whether you are meeting your performance objectives, use the qos statistics commands to monitor policy group and workload performance.
    You should monitor performance from the cluster. You should not use a tool on the host to monitor performance.
    Example
    The following command shows the performance of policy groups.
    cluster1::> qos statistics performance show
    Policy Group           IOPS      Throughput   Latency
    -------------------- -------- --------------- ----------
    -total-                 12316       47.76MB/s  1264.00us
    pg_app2                  7216       28.19MB/s   420.00us
    pg_vs1                   5008       19.56MB/s     2.45ms
    _System-Best-Effort        62       13.36KB/s     4.13ms
    _System-Background         30           0KB/s        0ms
    Example
    The following command shows the performance of workloads.
    cluster1::> qos statistics workload performance show
    Workload          ID     IOPS      Throughput    Latency
    --------------- ------ -------- ---------------- ----------
    -total-              -    12320        47.84MB/s  1215.00us
    app2-wid7967      7967     7219        28.20MB/s   319.00us
    vs1-wid12279     12279     5026        19.63MB/s     2.52ms
    _USERSPACE_APPS     14       55        10.92KB/s   236.00us
    _Scan_Backgro..   5688       20            0KB/s        0ms
  5. If necessary, use the qos policy-group modify command to adjust the policy group's maximum throughput limit.
    Example
    The following command modifies the maximum throughput for policy group pg-app2 to 20 MB/s.
    cluster1::> qos policy-group modify pg-app2 -max-throughput 20mb/s