ANF breakthrough mode volumes thoruput limit metric issue

Question

ANF breakthrough mode volumes thoruput limit metric issue

Srikanth L 95 Microsoft Employee

We recently deployed an ANF breakthrough mode volume. Post-deployment metrics indicate that the throughput limit was reached multiple times for this volume. The expectation with ANF breakthrough mode volumes is to deliver higher throughput compared to traditional large volumes. However, this appears to have been a bug, which was subsequently fixed by the SRE team on the backend. I would like to open a case with SRE to ensure this issue is formally documented as a bug and to track its resolution.

Praveen Bandaru 11,720 Reputation points Microsoft External Staff Moderator

2026-07-06T05:00:07.36+00:00
Hello Srikanth L

According to the docs, if a volume shows “Throughput limit reached”—meaning it is being throttled—the first thing to review is throughput sizing, service level, and whether the workload matches the volume. The documentation says that hitting the throughput limit usually means the volume is not sized appropriately for the application’s demands, and recommends increasing the limits by resizing or moving to a different service level.

First confirm that throughput throttling is actually happening

In Azure portal: ANF account → Volumes → (select volume) → Monitoring → Metrics

Check the “Throughput limit reached” metric; it indicates whether the throughput limit was reached and throttling occurred.

The docs also explain that this metric is evaluated in 5-minute intervals, and it will show 1 for the final minute of an interval if the limit was reached; if not, it will show 0 in the next interval.

If throttling is confirmed, increase the available throughput The recommended fixes in the metrics guidance are:

Resize the volume (auto QoS): increasing the volume size allocates more throughput and helps avoid throttling.

Change the service level: moving the volume/capacity pool to Premium or Ultra increases the throughput limits.

Adjust workload or mount options if needed: for NFS on Linux, the docs recommend tuning mount options such as using nconnect and keeping rsize/wsize at or below 256KiB.

Overprovision the QoS quota (if you use auto QoS): if the workload is throughput-bound, increasing the automatic QoS quota raises the throughput level.

If the higher performance is only needed temporarily, you can increase or decrease the quota dynamically without interrupting access.

Use the performance benchmarks as a Linux sanity check The benchmark document includes example expectations for large volume breakthrough mode on Linux in a specific scale-out test setup:

Example sequential read throughput: 50,000 MiB/s

Example sequential write throughput: 21,202 MiB/s

Example random read IOs and write IOs are also listed in the benchmark article.

Check the below documents for more understanig:
https://learn.microsoft.com/en-us/azure/azure-netapp-files/performance-large-volume-breakthrough-mode-linux

https://learn.microsoft.com/en-us/azure/azure-netapp-files/azure-netapp-files-service-levels

https://learn.microsoft.com/en-us/azure/azure-netapp-files/azure-netapp-files-metrics
Srikanth L 95 Reputation points Microsoft Employee

2026-07-06T05:05:22.0233333+00:00

Thanks for the details, if you are looking at the summary i have reported , this is already resolved by SRE and issue was flagged by SRE as bug. I want to have a case open with SRE to document the same and to have permanent fix for new breakthrough mode volumes

1 answer

Your answer

Srikanth L 95 Reputation points Microsoft Employee

2026-07-06T05:05:22.0233333+00:00

Thanks for the details, if you are looking at the summary i have reported , this is already resolved by SRE and issue was flagged by SRE as bug. I want to have a case open with SRE to document the same and to have permanent fix for new breakthrough mode volumes

Answer 1

Hello Srikanth L

After deploying an ANF breakthrough mode volume, the post-deployment metrics showed that the throughput limit was reached several times for this volume.

ANF breakthrough mode volumes are expected to provide higher throughput than traditional large volumes.

For this issue, we reached out to the backend team internally, and they fixed it. Actually, one ongoing issue occurring due to that problem you faced this issue. Finally, your issue was fixed.

Azure NetApp Files: https://feedback.azure.com/d365community/forum/7448e9b2-2225-ec11-b6e6-000d3a4f09d0

Azure VMware Solution: https://learn.microsoft.com/en-us/azure/azure-netapp-files/azure-netapp-files-service-levels