After you provision your Google Cloud Hyperdisk volumes, your application and operating systemmight require performance tuning to meet your performance needs.

In the following sections, we describe a few key elements that can be tuned forbetter performance and how you can apply some of these elements to specifictypes of workloads.

For an overview of how Google Cloud Hyperdisk performance works, seeAbout Hyperdisk performance.

Use a high I/O queue depth

Hyperdisk volumes have higher latency than locally attached disks such as localSSDs because they are network-attached devices. They can provide very high IOPSand throughput, but you need to make sure that enough I/O requests are done inparallel. The number of I/O requests done in parallel is referred to as theI/Oqueue depth.

The following tables show the recommended I/O queue depth to ensure you canachieve a certain performance level. The tables use a slightoverestimate of typical latency in order to show conservativerecommendations. The example assumes that you are using an I/O size of 16 KB.

Ensure you have free CPUs

Reading and writing to Hyperdisk volumes requires CPU cycles from your VM. Ifyour VM instance is starved for CPU, your application won't be able to managethe IOPS described earlier. To achieve very high, consistent IOPS levels, youmust have CPUs free to process I/O.

Review Hyperdisk performance metrics

You can review disk performance metrics inCloud Monitoring,Google Cloud's integrated monitoring solution. You can use these metricsto observe the performance of your disks and other VM resources under differentapplication workloads.

To learn more, seeReviewing disk performance metrics.

You can also use theObservability page in the console toview the diskperformance metrics.

What's next

• Learn aboutHyperdisk pricing.
• Analyze the provisioned IOPS for Hyperdisk volumes.