Edit

Elastic SAN Next Generation metrics

Azure provides metrics in the Azure portal that give you insight into your Azure Elastic SAN Next Generation resources. This article provides definitions of specific metrics you can select to monitor, their recommended use, how to view them at the SAN resource level, and how to calculate IOPS and throughput.

Important

Azure Elastic SAN Next Generation is currently in preview. See the Supplemental Terms of Use for Microsoft Azure Previews for legal terms that apply to Azure features that are in beta, preview, or otherwise not yet released into general availability.

View Elastic SAN Next Generation metrics

  1. In the Azure portal, open your Elastic SAN Next Generation resource.

  2. Under Monitoring, select Metrics.

  3. Select the Metric namespace and Metric.

  4. Select an aggregation.

  5. When supported by the selected metric, use available dimensions to filter the results.

Note

For Elastic SAN Next Generation, capacity metrics are available at the SAN resource level, with Volume Group and Volume dimensions that allow you to filter and analyze capacity usage for individual volume groups and volumes. Not all metrics are available at every resource scope. Metric availability and supported dimensions depend on the metric you select.

Available metrics

Elastic SAN Next Generation metrics are grouped into two categories:

  • Capacity metrics describe provisioned volume size, used volume size, and snapshot size.
  • Performance metrics describe service availability, I/O requests, data transfer, and request latency.

Capacity metrics definitions

Elastic SAN Next Generation exposes three capacity metrics: Provisioned Size, Used Size, and Snapshot Size. These metrics show the capacity allocated to volumes, the capacity used by volume data excluding snapshots, and the capacity used by snapshots, respectively. Use the Volume Group and Volume dimensions to filter results for individual volume groups and volumes.

Metric Definition Scope Recommended use
Provisioned Size The provisioned size of Elastic SAN Next Generation volumes in bytes. SAN resource; Volume Group and Volume dimensions Understand how much capacity you allocated to volumes.
Used Size The measured used size of Elastic SAN Next Generation volumes in bytes, excluding snapshots. SAN resource; Volume Group and Volume dimensions Monitor actual used size excluding snapshots.
Snapshot Size The measured size of snapshots of Elastic SAN Next Generation volumes in bytes. SAN resource; Volume Group and Volume dimensions Monitor snapshot consumption separately from used volume size.

Performance metrics definitions

Performance metrics describe the requests and I/O activity that Elastic SAN Next Generation processes. This category includes service availability, request counts, latency, and data transfer.

For Elastic SAN Next Generation, supported performance metrics can be viewed at the SAN resource level and analyzed using Volume Group and Volume dimensions.

Metric Unit Definition Recommended use
Availability Percent The percentage of availability for the storage service or specified API operation. Monitor service availability and configure availability alerts.
Transactions Count The number of requests made to the storage service or specified API operation. This count includes successful requests, failed requests, and requests that return errors. Understand I/O activity and calculate average IOPS over a selected interval.
E2E Latency Milliseconds The average end-to-end latency of successful requests made to the resource or specified API operation. Understand latency across the complete request path.
Server Latency Milliseconds The average time the service takes to process a successful request. This value doesn't include the network latency that's part of E2E Latency. Compare with E2E Latency to help understand the service-processing portion of latency.
Ingress Bytes The amount of data written to the SAN. Monitor write-oriented data transfer and calculate write throughput.
Egress Bytes The amount of data read from the SAN. Monitor read-oriented data transfer and calculate read throughput. This metric doesn't represent billable network egress.

Important

In Elastic SAN Next Generation metrics, Ingress represents data written to the SAN and Egress represents data read from the SAN. These definitions differ from how ingress and egress might be interpreted for Azure network traffic.

Calculate IOPS and throughput

Elastic SAN Next Generation provides foundational I/O telemetry through Transactions, Ingress, and Egress. You can use these metrics to calculate average IOPS and throughput for a selected time interval.

Calculate average IOPS

IOPS calculation:

  • Average IOPS = Total transactions in time bucket / Seconds in time bucket

  • Example: A one-minute bucket contains 120,000 transactions:

    120,000 / 60 = 2,000 average IOPS

To calculate IOPS in the Azure portal:

  1. Select Transactions.

  2. Set Aggregation to Total.

  3. Select a one-minute time granularity.

  4. Divide the displayed transaction value by 60.

Important

This calculation provides an average rate for the selected time bucket. It doesn't represent an instantaneous IOPS rate.

Calculate average throughput

Throughput calculation:

  • Average total throughput (MB/s) = (Ingress bytes + Egress bytes) / Seconds in time bucket / 1,000,000

For separate read and write views:

  • Average write throughput (MB/s) = Ingress bytes / Seconds in time bucket / 1,000,000
  • Average read throughput (MB/s) = Egress bytes / Seconds in time bucket / 1,000,000

Resource logging

Note

The resource logging guidance in this section applies only to General Purpose volume groups that use iSCSI. It doesn't apply to General Purpose or Business Critical volume groups that use direct attach.

You can configure the diagnostic settings of your Elastic SAN Next Generation to send Azure platform logs and metrics to supported destinations.

Elastic SAN Next Generation provides the Transactions log category for transaction logging. Available log categories include:

  • Write Success Requests
  • Write Failed Requests
  • Read Success Requests
  • Read Failed Requests
  • Persistent Reservation Requests
  • SendTargets Requests

Monitor workload performance

Monitor Availability

Use Availability to monitor the percentage of requests that the service successfully serves. For Elastic SAN Next Generation, use the Average aggregation to understand availability over the selected time period. You can also use minimum availability when configuring monitoring intended to identify periods of degraded availability.

How to create an alert for Availability < 99.9%

  1. Open the Create an alert rule dialog box. For more information, see Create or edit an alert rule.

  2. In the Scope tab, select your Elastic SAN Next Generation resource.

  3. In the Condition tab, select the Availability metric.

  4. In the Alert logic tab, select the following values:

    Field Description
    Threshold Static
    Aggregation type Average
    Operator Less than
    Threshold value 99.9
  5. Configure the evaluation frequency and lookback period appropriate for your monitoring scenario.

  6. Select Next to go to the Actions tab and add an action group. You can select an existing action group or create a new action group.

  7. Select Next to go to the Details tab and fill in the details of the alert, such as the alert name, description, and severity.

  8. Select Review + create to create the alert.

Next steps