Review Persistent Disk performance metrics

You can use disk metrics to observe your disks' performance and debug performance problems.

Disk metrics can help you answer questions such as the following:

  • What's the average read IOPS for a virtual machine (VM) instance's disks?
  • On average, what's the latency for read or write operations?
  • What's the average queue depth for a specific disk?

Review your disk's metrics to ensure its performance is sufficient for your workload. In addition, you should also do the following:

This document discusses the Persistent Disk metrics Compute Engine automatically collects from each VM and how to view them in Cloud Monitoring, which is Google Cloud's monitoring solution.

Available Persistent Disk metrics

You can view metrics in Cloud Monitoring, or programmatically retrieve Persistent Disk metrics using the REST API, client libraries, Metrics Query Language (MQL), and PromQL.

The following table lists the disk-specific metrics available for every disk. You can collect additional metrics if you install the Ops Agent on your VM.

For a full list of Compute Engine metrics, see Compute Engine metrics.

Each metric type in this table must be prefixed with compute.googleapis.com/, which has been omitted from the table for readability.

Display name
(Metric type)		 Description
Disk performance statusBETA
(instance/disk/disk_performance_status) 		The disk's health over the last minute. This metric indicates if the disk is performing normally or if its performance is affected by an incident within Compute Engine. Possible values are Healthy, Degraded, and Severely Degraded.
For more information, see Monitor a disk's health.
Average I/O latency
(instance/disk/average_io_latency) 		The disk's average read/write latency, in microseconds, for the last minute.
Average I/O queue depth
(instance/disk/average_io_queue_depth) 		The disk's average queue depth for read/write operations over the last minute.
Disk read bytes
(instance/disk/read_bytes_count) 		Average read throughput, or, the average number of bytes read or written over a period of time specified by the user*.
Disk write bytes
(instance/disk/write_bytes_count) 		Average write throughput, or, the average number of bytes written over a period of time specified by the user*.
Disk read operations
(instance/disk/read_ops_count) 		The average number of read operations over a period of time specified by the user*.
Disk write operations
(instance/disk/write_ops_count) 		The average number of write operations over a period of time specified by the user*.
Peak disk read bytes
(instance/disk/max_read_bytes_count) 		Peak read throughput, the maximum number of bytes read per second over a period of time specified by the user*.
Peak disk write bytes
(instance/disk/max_write_bytes_count) 		Peak write throughput, the maximum number of bytes written per second over a period of time specified by the user*.
Peak disk read ops
(instance/disk/max_read_ops_count) 		The maximum number of read operations per second over a period of time specified by the user*.
Peak disk write ops
(instance/disk/max_write_ops_count) 		The maximum number of read/write operations per second over a period of time specified by the user*.

* The period must be one minute or longer.

Visualize disk performance on a chart

You can visualize your disk's performance by plotting any of the metrics listed in the preceding section with Metrics Explorer. Metrics Explorer is part of Cloud Monitoring.

Example: Visualize average latency for the disks attached to a VM

To visualize the average latency for a VM's disks on a chart, follow these instructions. You can follow the same procedure for the other Persistent Disk metrics.

  1. In the Google Cloud console, go to the  Metrics explorer page:

    Go to Metrics explorer

    If you use the search bar to find this page, then select the result whose subheading is Monitoring.

  2. In the Metric element, expand the Select a metric menu, enter VM Instance in the filter bar, and then use the submenus to select a specific resource type and metric:
    1. In the Active resources menu, select VM Instance.
    2. In the Active metric categories menu, select Instance.
    3. In the Active metrics menu, select Disk average latency.
    4. Click Apply.
    The fully qualified name for this metric is compute.googleapis.com/instance/disk/average_io_latency.
  3. Configure how the data is viewed. To display only the metrics for each disk attached to a specific instance, follow these steps:
    • In the Filter element, click Add filter, and then select instance_name. For the value, select a specific instance name.
    • In the Aggregation entry, set the first menu to Mean and the second menu to device_name.

    For more information about configuring a chart, see Select metrics when using Metrics Explorer.

What's next