Cloud Pricing Comparison 2025: Compute, Storage, and Networking

As many as 90% of organizations have adopted the cloud in some capacity. Yet, more than 60% believe they are not fully realizing the cost benefits they had anticipated. Nearly 50% of businesses are struggling with their cloud bills, leading a growing number to consider full or partial repatriation. Many blame cloud’s spiralling costs on complex billing models and unpredictability that organizations fail to assess. As such, it makes sense to evaluate and compare cloud costs and pricing models before choosing a cloud service provider.

There are many factors that determine the overall cloud costs, some obvious, others not so much. In this article, we unpack all those factors and compare the pricing from top cloud service providers, so you can pick the most cost-effective option for your needs and avoid the dreaded cloud bill shock. Keep on reading for our ultimate cloud cost comparison.

Understanding Cloud Pricing Models

Cloud service providers offer several billing models to choose from.The best pricing model for your organization depends on factors like the type of workloads you run, their usage patterns and demand predictability.

1. Pay-As-You-Go (PAYG)/On-Demand: It’s the dominant billing model in the cloud. With this model, you only pay for the resources you consume, often charged on a per-second, per-minute, or a per-hour-basis. While this model gives you the most flexibility and reduces resource wastage, on-demand pricing can be much higher than other commitment or reservation-based pricing models. Still, it’s the most suitable model for unpredictable workloads and short-term projects, where it is difficult to predict the demand in advance.
2. Reserved Instance/Committed-Use Discount: Reserved instances are your best option when you can confidently predict the resource demand. CSPs often offer discounts up to 70% when you reserve a specific amount of compute power for a period of time—typically a 1-year or 3-year commitment. If usage exceeds your reserved capacity, you get charged at the standard PAYG rate for the excess resources consumed.
3. Savings Plans: Providers like AWS and Azure offer savings plans in addition to reserved instances (RI). Users can commit to spending a certain amount of money with the cloud service provider and get discounted rates in exchange. This model gives more flexibility than RI discounts as you don’t need to restrict yourself to specific types of compute instances within a pre-chosen location. You can choose any instance type, size, operating system, or tenancy and still get a discounted price for it. It’s like buying specific items at a discounted rate versus using a coupon that applies to your entire grocery bill.
4. Sustained-Use Discounts: GCP also offers sustained-use discounts, which are applied automatically once you’ve run your VMs for a certain amount of time. These discounts apply even if you haven’t reserved the instance or committed to usage beforehand.
5. Spot Instances: Spot instances are the spare capacity that CSPs have but isn’t being actively utilized at any given moment. CSPs offer these spot instances at steep discounts (up to 90%) to ensure maximum utilization. You can place a bid on available spot instances and the user with the highest bid secures the instance. However, the CSP can reclaim the instance on super short notice—typically just two minutes—if other higher-paying customers need it.
6. Subscription-Based Pricing: It’s the go-to pricing model for cloud services and SaaS offerings. Users pay a monthly or annual fee for specific cloud services or software.

How to Calculate and Compare Prices

You should consider and compare the cost of cloud resources and services for each provider to get a good idea of cost-effective options. However, to do that, you must first assess your needs and compare CSPs’ upfront prices as well as additional or hidden charges based on that.

1. Assess Your Needs

The cost of different resources and instance types is generally available on the pricing pages of CSPs. However, those prices can fluctuate based on factors like the region and running duration. Resources hosted in certain regions can be more expensive than others. Similarly, heavy or long-running workloads may be eligible for various discounts. Consider all such factors holistically and factor in applicable discounts when comparing cloud prices.

2. Consider Additional Costs

Don’t forget additional and hidden costs when comparing cloud providers. These can include data egress charges to the internet and across regions. CSPs offer different support levels at various monthly rates. So, consider if you need dedicated support from your CSP. You must also include the cost of other subscription-based services and software you need in your cloud pricing comparison.

To get you started, we have compared the prices for basic compute, storage, and networking across the top five cloud service providers, including AWS, Azure, GCP, Alibaba, and OCI. These cloud providers individually hold a significant share of the global cloud market and play a crucial role in shaping any organization's cloud strategy and service choices.

Cloud Pricing Comparison: Compute

The cost of compute instances depends on various factors, including the type and size of the instance, billing models, and the cloud region where the instance is deployed. Compute or memory-optimized instances tend to be more expensive than general-purpose instances. Similarly, on-demand instances incur higher costs compared to spot instances. Licensing costs and add-ons like AI/ML accelerators and managed services integrations can also affect your final cloud bills. However, those are some of the more complex cost factors that you should consider after you have a basic cloud pricing comparison down.

For the sake of this article, we first compare basic compute instances—the smallest and largest offered by the CSPs within their default region and a medium instance (2 vCPUs, 8 GB RAM) for consistency and to serve as a baseline. Within each size category, the highest price is highlighted in red, and the lowest in green.

‍Please note that these are estimated prices listed at the time of writing this article and may fluctuate over time. Actual prices can also vary based on promotions, region, and/or specific contract terms.

Comparison by Instance Size

Based on this basic comparison, Oracle comes out to be the most cost-effective in the standard, medium instance and large instance category. However, do consider that the large instances offered by AWS and Azure are considerably larger than Oracle’s.

The pay-as-you-go prices mentioned above go down substantially when you consider the various discounts CSPs offer. To factor in available discounts, we also compare the various discounted prices you can get for the standard medium instance.

Discounted Price Comparison

AWS (Amazon Web Services)

Instance Type: m5.large (2 vCPUs, 8 GB RAM)

Pricing Option: Price per Hour

On-Demand: $0.096

Reserved (1 Year): $0.060 (38% savings)

Reserved (3 Years): $0.041 (57% savings)

Savings Plan (1 Year): $0.071 (26% savings)

Savings Plan (1 Year): $0.049 (49% savings)

Spot Pricing: $0.0346 (63.96% savings)

Microsoft Azure

Instance Type: D2s_v3 (2 vCPUs,  8 GB RAM)

Pricing Option: Price per Hour

On-Demand: $0.0960

Reserved (1 Year): $0.0572(40% savings)

Reserved (3 Years): $0.0368(61% savings)

1 Year Savings Plan: $0.0662 (31% savings)

3 Year Savings Plan: $0.0451 (53% savings)

Spot: $0.0141 (85% savings)

Google Cloud Platform (GCP)

Instance Type: n2-standard-2 (2 vCPUs, 8 GB RAM)

Pricing Option: Price per Hour

On-Demand: $0.0985

Reserved (1 Year): $0.0625 (37% savings)

Reserved (3 Years): $0.045 (54% savings)

Sustained Use Pricing: $0.0788 (20% savings)

Preemptible VMs: $0.0274 (72% savings)

Alibaba Cloud

Instance Type: ecs.g5.large (2 vCPUs,  8 GB RAM)

Pricing Option: Price per Hour

On-Demand: $0.12

Discounted Price (Monthly Pay): $0.078 (35% savings)

Discounted Price (Yearly Pay): $0.066 (45% savings)

Oracle Cloud

Instance Type: VM.Standard.E4.Flex (2 vCPUs, 8 GB RAM)

Pricing Option: Price per Hour

On-Demand: $0.0377

Reserved Capacity (50% Unused): $0.035 (7% savings)

Preemptible: $0.0188 (50% savings)

Overall, AWS and Azure offer some of the highest percentage savings on long-term commitment pricing models and spot instances.

Cloud Pricing Comparison: Storage

Cloud storage costs depend on the storage type (block, object, and file) and size as well as the access frequency. Cloud providers also charge for the number of read and write operations you make to the storage. So, the more you interact with the data, the higher the cost. To give you a basic cloud price comparison for storage, we compare the monthly cost of block storage per GB of stored data, along with the maximum IOPS (Input/Output Operations Per Second) and throughput you can get with that storage.

AWS

Amazon Elastic Block Store (EBS) General Purpose SSD (gp3)

Price: $0.08/GB-month

3,000 IOPS free and $0.005/provisioned IOPS-month over 3,000;

125 MB/s free and $0.04/provisioned MB/s-month over 125.

Volume Size: 1 GB - 16 TB

Max IOPS/Volume: 16,000

Max Throughput/Volume: 1,000 MB/s

AWS Free Tier includes 30 GB of storage, 2 million I/Os, and 1 GB of snapshot storage with Amazon Elastic Block Store (EBS). The above pricing applies once you go beyond the free tier.

Azure

Azure Premium SSD v2

Price: $0.081/GiB/Month

Up to 3,000 IOPS at no additional cost and $0.0052/provisioned IOPS over 3,000 IOPS.

Up to 125 MB/s at no additional cost and $0.041/provisioned MB/s over 125 MB/s.

Disk Size: 1 GiB-64 TiBs

Max IOPS: 80,000

Max Throughput (MB/s): 1,200

Azure's Premium SSD v2 storage is not included in the free tier.

GCP  

Persistent Disk SSD

Price: $0.17/GB/month

Max IOPS: 100000 read and 30000 write IOPS for disks with 4 TB or more.

Max Throughput: Up to 1200 MB/s for read and 1000 MB/s write operations

Free tier includes 5 GB-months of regional storage only available in us-east1, us-west1, and us-central1 regions.

Alibaba

Standard SSD

Price: $0.187 GiB/Month

Max IOPS: 25,000

Max Throughput: 300MB

Oracle (OCI)

Block Volumes (SSD)

Price: $0.0255/GB/month

Max IOPS: 18000

Max Throughput: 140 MBps

Oracle Always Free resources include a total of 200 GB of Block Volume storage and 5 volume backups.

Again, OCI offers some of the most cost-effective storage solutions among the top 5 cloud service providers.

Cloud Pricing Comparison: Networking

Networking costs in cloud computing generally include data transfer costs to and from the cloud, cloud regions, and availability zones as well as additional costs for service-to-service communications and security features. For this pricing comparison, we consider the cost of data transfer:

• IN to the cloud: Data ingress
• OUT to the internet: Internet egress
• Between different cloud regions: Cross-region data transfer

AWS

Free Tier: Data transfer into AWS from the internet is free. For data egress, AWS customers get 100GB of data transfer out to the internet for free each month.

Internet Egress:

• First 10 TB / Month: $0.09 per GB
• Next 40 TB / Month: $0.085 per GB
• Next 100 TB / Month: $0.07 per GB
• Greater than 150 TB / Month: $0.05 per GB

Cross-Region Data Transfer: $0.01 to $0.09 per GB, depending on the regions.

Azure

Free Tier: All data transfer into Azure from the internet is free. Data transfers within the same AZs (Availability Zones) are also free. Azure customers also get 100 GB/month of egress data for free.

Internet Egress:

• First 100 GB/month: Free
• Next 10 TB/month: $0.08 per GB
• Next 40 TB/month: $0.065 per GB
• Next 100 TB/month: $0.06 per GB
• Next 350 TB/month: $0.04 per GB

These costs are for traffic originating from North America and Europe and can vary depending on the source traffic geolocation.

Cross-Region Data Transfer: $0.02 to  $0.16 per GB

GCP

Free Tier: Data transferred into GCP is free of charge. Data transfers within the same Google Cloud zones are also free of charge.

Internet Egress:

• 0-200 GiB/month: No charge
• 200 GiB-10 TiB/month: $0.085 per GB
• 10-150 TiB/month: $0.065 per GB
• 150-500 TiB/month: $0.045 per GB

Pricing is based on source geolocation of traffic.

Intra-Region Data Transfer: $0.01 per GB

Cross-Region Data Transfer: $0.02 to $0.14 per GB, depending on the source and destination location.

Alibaba

Free Tier: The first 200 GB of internet data transfers per month is free of charge.

Internet Egress:

Cross-Region Data Transfer: $0.02/GB to $0.80/GB

Oracle OCI

Free Tier: Like other cloud providers, inbound data transfer is free. Outbound data transfer is free for the first 10 TB of data.

Internet Egress:

Over 10 TB / Month:

• $0.0085 per GB for data originating in North America, Europe, and the UK.
• $0.025 per GB for data originating in APAC, Japan, and South America.
• $0.05 per GB for data originating in Middle East and Africa

How to implement Cloud Optimization Strategies to Reduce Your Cloud Bill

Without an overarching cloud strategy, cloud costs can escalate out of control. As you can see in our cloud pricing comparison, the differences in pricing across CSPs are often minimal. Even if you opt for the most cost-effective option, you could still end up spending more than the pricier alternatives if you don’t optimize your cloud spending with a few, simple strategies:

• Avoid Overspending: More than 90% of organizations report waste in their cloud spend. You should implement best practices to optimize cloud resource consumption and take precautions to avoid overspending. These basic strategies can go a long way:
  
  1. Set up cloud budgets and actively monitor cloud spend.
  2. Monitor cloud resource consumption.
  3. Right-size cloud resources to match workload demands.
  4. Spot and terminate idle resources.
  5. Use auto-scaling to prevent over-provisioning during off-peak demand.
• Perform Continuous Optimization: Cloud pricing and discounts continuously fluctuate and so do your workload demands. Make sure you continuously evaluate your needs and cloud offerings to make use of reserved instance discounts optimally. Also, keep an eye on the available spot instances for substantial savings where possible.
• Adopt a Multi-Cloud Strategy: As shown in our price comparison, different CSPs have best-pricing across various service categories and instance families. Instead of selecting one cloud provider, you can choose multiple cloud providers and host each workload in the optimal environment for its cost and performance requirements. A multi-cloud set up will also allow you to maximize savings by choosing the cloud and region offering the lowest spot instance pricing at any given moment.

• Use Cost Monitoring Tools: As cloud environments get complex, it can be difficult to keep tabs on all the different factors contributing to the cloud bill. Cloud cost monitoring tools can help provide a clear picture of historical and current cloud costs along with future forecasts. You can choose vendor-specific tools like AWS Cost Explorer, Azure Cost Management, and GCP Cloud Billing, or a vendor-neutral, comprehensive cloud management platform like emma to closely track and optimize your cloud spend.
• Consider Specialized CSP Offerings: Certain CSPs have specialized offerings for specific workloads and environments. For instance, you can achieve greater cost efficiency with Azure or Oracle if you already have enterprise licensing for their products. CSPs also have a different pricing structure for hybrid cloud needs. Check out our comprehensive review of top 10 CSPs and hybrid cloud provider comparison to better understand your needs and evaluate your options accordingly.

How to Manage and Control Cross-Cloud Costs

emma cloud management platform empowers organizations to take control of their cloud landscape, even if it spans on-premise and multiple clouds. It’s a no-code, vendor-agnostic cloud management platform, which means teams with minimal cloud expertise can use it to find and select optimal cloud environments for their applications and workloads.

emma’s AI powered recommendations can reduce your cloud bill by up to 75%. Here are some amazing features that can help you and your teams monitor, evaluate, and optimize cloud spend:

1. Find and utilize spot instances in real-time and across multiple cloud providers and services.
2. Monitor resource utilization across projects and cloud providers and use those insights for rightsizing and to identify idle instances.
3. Access real-time and historical cost and usage insights as well as demand forecasts to make the best use of reserved instance discounts across cloud providers.
4. Set cost limits per project and cloud service provider, and receive alerts when you’re about to reach the threshold.
5. Get real-time recommendations as more suitable and cost-effective options become available. Seamlessly implement emma’s recommendations by moving existing workloads across clouds as and when needed.
6. Minimize cross-region or cloud-to-cloud data transfer costs by up to 80% using emma’s egress cost optimizer and emma’s high-performance multi-cloud networking backbone.

Try emma today with a risk-free, 14-day trial—no commitments required!