Does Disabling Azure Front Door Caching Reduce Premium Data Transfer Out Cost?

Question

Does Disabling Azure Front Door Caching Reduce Premium Data Transfer Out Cost?

Supakrit (Tan) Aphonmaeklong 20

Our Azure Front Door Premium profile, afd-ui-prod, shows approximately 3.45 TB of data transfer, with 99.65% originating from Thailand.

Azure Front Door analytics indicate that almost all transferred data comes from clients located in Thailand, while our actual production customers are primarily located in Europe. We believe most of the Thailand traffic is generated by our internal development teams rather than external production users.

For the reviewed period, Thailand traffic shows:

Requests: 5,869,306

Request share: 97.58%

Cache hit ratio: 88.92%

Data transferred: 3,446,396,658,467 bytes, approximately 3.45 TB

Data transfer share: 99.65%

HTTP 4xx responses: 460

HTTP 5xx responses: 2

The billing meter is Premium Data Transfer Out, with a cost of approximately USD 1,130.

Please confirm whether disabling Azure Front Door edge caching would reduce this charge. Our understanding is that Front Door bills for bytes sent from the edge to clients regardless of whether the response is served from cache. Therefore, disabling caching would not reduce the Premium Data Transfer Out cost and may instead increase traffic and workload on the origin.

Shree Hima Bindu Maganti 7,590 Reputation points Microsoft External Staff Moderator

2026-08-14T15:31:10.54+00:00

Hi @Supakrit (Tan) Aphonmaeklong

Thank you for sharing the detailed traffic and billing information.

Based on the details provided, disabling Azure Front Door caching would not reduce the Premium Data Transfer Out charges. The data transfer out charge is based on the data delivered from the Azure Front Door edge to the client, regardless of whether the response is served from cache or retrieved from the origin.

In fact, with the current cache hit ratio of approximately 88.92%, caching is helping reduce requests and data transfer between Azure Front Door and the origin. Disabling caching could increase the load and traffic to the origin without reducing the edge-to-client data transfer charges.

Since approximately 99.65% of the data transfer is associated with traffic from Thailand, we recommend investigating the source of this traffic further, particularly whether it is generated by internal development users. Reviewing the source IPs, URLs, and traffic patterns would help identify the appropriate way to control or reduce this traffic.

Therefore, we would not recommend disabling caching as a cost-optimization measure. Instead, we recommend focusing on identifying and controlling the unexpected Thailand traffic.
https://learn.microsoft.com/en-us/azure/frontdoor/billing
https://learn.microsoft.com/en-us/azure/frontdoor/standard-premium/how-to-reports?tabs=traffic-by-domain
Please let us know if you have any further questions or if you would like us to clarify any of the above.

1 answer

Your answer

Shree Hima Bindu Maganti 7,590 Reputation points Microsoft External Staff Moderator

2026-08-14T15:31:10.54+00:00

Hi @Supakrit (Tan) Aphonmaeklong

Thank you for sharing the detailed traffic and billing information.

Based on the details provided, disabling Azure Front Door caching would not reduce the Premium Data Transfer Out charges. The data transfer out charge is based on the data delivered from the Azure Front Door edge to the client, regardless of whether the response is served from cache or retrieved from the origin.

In fact, with the current cache hit ratio of approximately 88.92%, caching is helping reduce requests and data transfer between Azure Front Door and the origin. Disabling caching could increase the load and traffic to the origin without reducing the edge-to-client data transfer charges.

Since approximately 99.65% of the data transfer is associated with traffic from Thailand, we recommend investigating the source of this traffic further, particularly whether it is generated by internal development users. Reviewing the source IPs, URLs, and traffic patterns would help identify the appropriate way to control or reduce this traffic.

Therefore, we would not recommend disabling caching as a cost-optimization measure. Instead, we recommend focusing on identifying and controlling the unexpected Thailand traffic.
https://learn.microsoft.com/en-us/azure/frontdoor/billing
https://learn.microsoft.com/en-us/azure/frontdoor/standard-premium/how-to-reports?tabs=traffic-by-domain
Please let us know if you have any further questions or if you would like us to clarify any of the above.

Answer 1

Yes. Disabling Azure Front Door caching would not reduce the Premium Data Transfer Out charge, because that charge is based on data sent from Azure Front Door edge to clients.

The supported cost guidance shows:

Caching reduces calls from Azure Front Door to the origin, which optimizes edge-to-origin data transfer costs.
Compression reduces bandwidth consumption.
Front Door cost assessment treats egress from Azure Front Door edge to client as a separate billing dimension.

For this scenario, the Thailand traffic already has an 88.92% cache hit ratio, which means caching is currently reducing origin fetches. If caching is disabled:

Edge-to-client transfer would still occur for the same client downloads.
Origin traffic would likely increase, because more requests would be forwarded to the origin.
Origin workload would likely increase.
Overall cost could increase in other meters, while the edge-to-client transfer charge would not be reduced just because caching was turned off.

If the goal is cost reduction, the supported options are:

Keep caching enabled for content that does not change often.
Enable compression where supported.
Use WAF controls such as rate limiting or geo-filtering to restrict unnecessary incoming traffic.
Review whether the Thailand traffic should be allowed at all, since WAF rules can restrict incoming traffic and help optimize costs.

References: