# Test Azure storage bursts after credit exhaustion, not only at startup

> How should a performance test distinguish initial burst credits from sustained Azure VM and disk capability?

- Canonical URL: https://update.dsesecurity.com/updates/dse-20260909-046-test-azure-storage-bursts-after-credit-exhaustion-not-only-at-startup/
- Publisher: Detection Systems & Engineering (DSE Security)
- Author: DSE Security Editorial Team
- Published: 2026-09-10T00:31:10+00:00
- Modified: 2026-09-10T00:32:01+00:00
- Last reviewed by DSE: 2026-09-09
- Resource type: Guide
- DSE priority: Information
- Topics: IT, Networks & Infrastructure
- Reading time: 2 minutes

## What you need to know

How should a performance test distinguish initial burst credits from sustained Azure VM and disk capability?

## Potentially affected

Teams testing Azure VMs and managed disks that support credit-based storage bursting.

## DSE recommendation

Include both a fresh-credit run and a sustained-demand run in the storage performance acceptance test.

## Article

## Source facts

Azure VM and disk bursting operate independently. In the credit-based model, each resource has its own credits, which accumulate when IOPS or bandwidth use is below its performance target. IOPS and bandwidth credits are tracked separately, so one can accumulate while the other is being spent. Microsoft describes credit-based disk bursting as best effort, not guaranteed. [Microsoft Learn](https://learn.microsoft.com/en-us/azure/virtual-machines/disk-bursting).

## Applicability

Use this check only after confirming that the chosen VM and disk sizes support the relevant bursting model. Separate credit-based behavior from on-demand disk bursting. Record the test’s starting conditions and workload duration instead of comparing results from an idle resource with results from an already busy one.

## DSE recommendation

Include both a fresh-credit run and a sustained-demand run in the storage performance acceptance test. Ask the application owner to identify the longest expected busy period and the acceptable behavior after initial headroom is consumed. Evaluate VM-level and disk-level constraints separately. Keep any brief startup improvement distinct from the sustained result used to approve the production workload.

## Verification

Capture IOPS, bandwidth, and available bursting observations over the complete trial rather than retaining only the fastest interval. Compare the later workload behavior with the application’s acceptance criteria and the documented baseline targets. If performance changes as credits are consumed, record that transition rather than averaging it away. Repeat the test from a documented starting condition before attributing differences to a configuration change.

## Official references

[Microsoft Learn: Managed disk bursting](https://learn.microsoft.com/en-us/azure/virtual-machines/disk-bursting). Source reviewed September 9, 2026.

## Primary reference

- Name: Managed disk bursting - Azure Virtual Machines | Microsoft Learn
- Authority: Microsoft Learn
- URL: https://learn.microsoft.com/en-us/azure/virtual-machines/disk-bursting
- Source publication date: Not stated by the source

## Citation and use

Preferred citation: “Test Azure storage bursts after credit exhaustion, not only at startup,” DSE Security, https://update.dsesecurity.com/updates/dse-20260909-046-test-azure-storage-bursts-after-credit-exhaustion-not-only-at-startup/
Publishing principles: https://update.dsesecurity.com/updates/dse-updates-editorial-methodology/
Usage and citation policy: https://update.dsesecurity.com/usage/
Copyright © 2026 Detection Systems & Engineering. All rights reserved.