Comparing Azure vs AWS pricing has always been difficult due to the frequency with which prices change. However, it doesn’t have to be an “all or nothing” choice between the two cloud service providers. You can pick which services you use from each, provided you maintain control of your multicloud environment.
Azure vs AWS pricing comparisons are not only complicated by frequent price changes and new services. Improvements to existing services (for example, latest generation instances/VMs) also complicate comparisons because they tend to provide better performance for the same price or less. Similarly, the availability of a service, the region(s) in which the service is deployed, and the committed use or committed payment discount applied to the service can also make a difference.
On Demand Azure vs AWS pricing
On Demand Azure vs AWS pricing comparisons are the simplest to make. In the following tables, we've taken the most popular sizes of AWS instances and Azure Virtual Machines (VMs) and compared their prices for both Linux OS and Windows OS with standard SQL. Prices are per hour for the latest generation AWS instances and Azure VMs deployed in the US East region.
On-Demand Azure vs AWS Pricing - Linux OS (October 2020)
On-Demand Azure vs AWS Pricing - Windows OS (October 2020)
What you also have to consider is that both Azure and AWS charge for outbound data transfers, block storage, load balancing, additional IP addresses, and premium support. Of these services, Azure is marginally cheaper for outbound data transfers and premium support at the Enterprise level. However, AWS' default EBS block storage prices are much less expensive than Azure's equivalent, and AWS also beats Azure on the cost of its load balancing services (correct as of October 2020 in the US East region).
You can see more detailed information on how AWS charges for data requests, transfers, and retrievals in our eBook: The Ultimate Guide to Amazon Cloud Storage Pricing
How committed use and committed spend discounts work
Both Azure and AWS offer discounted pricing when businesses commit to a predetermined level of usage over a fixed term, known as "Reservations" or "Reserved Instances."
Azure Reservations can apply to compute, storage, or app services, and AWS offers Reserved Instances (RIs) for EC2, RDS, ElastiCache, Elasticsearch, Redshift, and DynamoDB. In addition to Reserved Instances, AWS also offers Savings Plans for EC2, Fargate, and Lambda, where you’re committing to a minimum dollar spend per hour. The options available to businesses include:
- Commit to a one-year term or three-year term (AWS and Azure)
- Pay the full amount upfront to obtain the maximum discount (AWS only)
- Pay a partial amount upfront and the balance monthly (AWS only)
- Pay monthly for the term of the purchase (AWS and Azure - Since September 2019, it's been possible to pay monthly for Azure RIs and RIs purchased at no additional cost/discount)
If your business' circumstances change during the term of a purchase, AWS hosts a marketplace for buying and selling Reserved Instances. Azure offers an exchange or refund service (pro-rata less an administration fee), or you can take advantage of AWS' Convertible Reserved Instances that allow you to change most family types during the term of your purchase for a smaller discount on the hourly rate. (Azure allows you to exchange VM sizes, but only within the same family type and region).
As an alternative to Reserved Instances, which commit businesses to a predetermined level of usage, AWS introduced Savings Plans in November 2019, where businesses now have the option of committing to a predetermined level of spend.
With regard to an Azure vs AWS pricing comparison for committed use reservations (Azure doesn't currently offer a discount program similar to AWS Savings Plans), the percentage discount you receive from each provider varies according to the type of VM or instance you commit to, as well as the length of the time committment, and how much you pay upfront. The tables below illustrate some of the Reserved Instance discounts that can be achieved based on the instances and VMs mentioned above.
And for in-depth comparison of discounts, commitments, and reservations between AWS, Azure, and Google Cloud, see our article here.
Azure vs AWS Pricing - Reserved Instance Discount Percentages
|Azure - 3 year term||62%||63%||62%|
|Azure - 1 year term||41%||41%||41%|
|AWS - 3 year term||62%||62%||62%|
|AWS - 1 year term||41%||41%||41%|
|AWS - Convertible 3 year||54%||56%||53%|
|AWS - Convertible 1 year||31%||32%||33%|
An important point to consider about Standard Reserved Instances is that, once you commit to a one or three-year term, you're locked into the commitment at the price at the time. If there's a price cut during the term of the commitment (historically AWS has cut its prices by an average of 8.1% each year), you won't benefit from the price reduction. If you opt instead for Convertible Reserved Instances or AWS Savings Plans, you can benefit from subsequent price reductions.
A second important note to take into consideration for businesses who've paid for Windows Server and SQL Server licenses, is that Azure runs a Hybrid Benefit discount program, which increases the savings on Azure Reserved Instances by an average additional 8% when you bring your Windows Server and SQL Server on-premises licenses to Azure.
Azure vs AWS pricing for data storage
Although the majority of cloud computing budgets is allocated to instances/VMs, there are many other service prices businesses should compare to ensure they're getting the best bang for their buck. Storage is just one of many services businesses are likely to use when taking advantage of the agility, flexibility, and scalability of cloud computing, and costs can vary just as much as with instances and VMs.
The table below illustrates the differences between Azure vs AWS pricing for storing frequently accessed data, infrequently accessed data, and archive data. Charges may apply for retrieving data from infrequently accessed, glacier, and archive storage, but price differences between the two cloud service providers are negligible (although large volume businesses should conduct their own comparisons).
Azure vs AWS Pricing - Data Storage Prices (October 2020)
|First 50TB per month||$0.0208/GB||$0.0230/GB|
|Next 450TB per month||$0.0200/GB||$0.0220/GB|
|Over 500TB per month||$0.0192/GB||$0.0210/GB|
|One Zone Infrequent Access||n/a||$0.0100/GB|
Other Azure and AWS services to consider
Among the other services for which an Azure vs AWS pricing comparison can prove valuable, we suggest comparing the relative costs of deploying containers and the relative costs of off-peak batch processing. Both Azure and AWS provide two types of container services, plus an array of container registries, container orchestrators, and serverless compute engines for containers.
You can see an in-depth comparison of AWS, Azure, and Google Cloud container services in our article here.
With regards to off-peak batch processing services, both Azure and AWS now call their service “spot instances/VMs” (previously Azure's service was called Low Priority VMs). Both services work in the same way, as you state how much you're willing to pay for the providers' excess capacities, and provided excess capacity is available at the price you want, you can save up to 90% against On Demand prices.
AWS also offers an Intelligent Tiering service that transfers data between Frequent Access, Infrequent Access, and most recently announced at AWS re:Invent 2020, Archive and Deep Archive storage tiers, in exchange for a small charge, without impacting performance.
Managing different cloud providers, services, and prices
It was mentioned at the beginning of this article that the objective of an Azure vs AWS pricing comparison is not necessarily to make a decision between which of the cloud service providers you should use exclusively, but to identify where costs can be saved by using some services provided by Azure and other services provided by AWS. However, reducing costs in this way can have a trade-off—cloud management.
Managing one account with one cloud service provider can be hard enough, but when you're managing two or more accounts with different cloud service providers, the complexity of controlling a multicloud environment can increase out of proportion. For this reason, it's recommended that businesses taking advantage of a multicloud strategy, implement a cloud financial management practice.
Cloud Financial Management (CFM), also known as FinOps or Cloud Cost Management, is a function that helps align and develop financial goals, drive a cost-conscious culture, establish guardrails to meet financial targets, and gain greater business efficiencies.
And for more insights into how AWS and Azure compare, see our in-depth eBook: Comparing Services For The Big Three Cloud Providers