How Do Cloud Server Costs Work? A Quick Cost Comparison

Written by Keren Shmuely

What Are Cloud Servers? 

A cloud server is a centralized server resource hosted and served over a network (usually the Internet) and accessed by multiple users on an as-needed basis. 

Cloud servers perform all the same functions as traditional on-premise servers, providing processing power, storage, networking, and the ability to run applications. They are typically managed by virtualization platforms, which enable multiple virtual servers to run on one physical machine in a cloud data center.

Cloud servers can be located anywhere in the world, and can provide services remotely through cloud computing environments. In contrast, traditional dedicated server hardware is often installed on-premises and used by only one organization.

This is part of a series of articles about cloud cost management.

In this article:

Cloud Server Deployment Models

Cloud servers virtualize physical servers that users can access from remote locations. Server virtualization is usually done via a hypervisor. The virtualization platform creates a virtual cloud server, powered by the computing resources of a physical server running in the cloud provider’s data center. Organizations can access these virtual servers from any physical location via the Internet (if access is allowed). Cloud servers are configured and managed through a web-based console or API.

In a public cloud computing environment, the cloud provider typically provides access to virtual servers, storage, and other services and resources for a fee under a pay-as-you-go subscription model. There are three common deployment models:

  • Infrastructure as a Service (IaaS)—a cloud deployment model that includes traditional infrastructure like virtual servers, networking, and storage.
  • Platform as a Service (PaaS)—provides customers with a complete cloud environment and hardware and software tools to develop applications. These tools are based on cloud server, storage and network resources. 
  • Software as a Service (SaaS)—vendors deliver fully managed software products to customers via the cloud. SaaS applications get their computing resources from cloud servers.

Private cloud servers work in a similar way, but the physical servers belong to the company’s privately owned infrastructure, and are not shared by other organizations.

New call-to-action

Cloud-Based Server Costs: Comparing the Big Three Cloud Providers

Let’s get down to details and see how cloud servers are priced, when offered in an IaaS model, across the big three cloud providers: Amazon, Azure, and Google Cloud. A complete comparison across each of the hundreds of instance types is outside the scope of this article. Instead, we compare two groups of similar instances across the three providers, to give a taste of actual pricing differences.

Keep in mind that Amazon, Azure, and Google Cloud pricing is subject to change – for the latest pricing please see each provider’s official pricing page.

Which Instances are We Comparing?

To ensure a fair comparison, we selected instances with similar parameters across all providers:

  • Instances running in the US East region
  • Instances with the Linux operating system
  • Instances with 4 vCPUs

On the memory side, we’ll compare two groups of instances:

  • General purpose instances with 8 GB of RAM (except for Google Cloud which provides 16 GB even for this category)
  • Compute optimized instances with 16 GB of RAM

Lastly, we’ll compare three pricing models:

  • On-demand pricing
  • Reserved instance with 1 year upfront payment
  • Spot instance

Comparing General Purpose Instances

Cloud providerInstance typeOn-Demand PriceReserved Price (1 Year Upfront)*Spot Price*
AWSt4g.xlarge$0.1344$0.079 (41%)$0.0403 (30%)
AzureB4ms$0.166$0.0974 (41%)$0.0638 (67%)
Google Cloud Platforme2-standard-4$0.150924$0.0951 (63%)$0.0453 (70%)

* The percentage in brackets is the discount offered compared to the on-demand price.

Amazon is cheapest for on-demand instances, while Google Cloud consistently offers the biggest discounts for reserved and spot prices. 

Google Cloud’s price is competitive when considering that its instance offers 16 GB of RAM, more than the comparable AWS and Azure instances.

Comparing Compute Optimized Instances

Cloud providerInstance typeOn-Demand PriceReserved Price (1 Year Upfront)*Spot Price*
AWSc6a.xlarge$0.153$0.094 (38%)$0.068 (44%)
AzureF4s v2$0.1690$0.10 (41%)$0.0259 (85%)
Google Cloud Platformc2-standard-4$0.2351$0.1316 (63%)$0.0540 (77%)

* The percentage in brackets is the discount offered compared to the on-demand price.

Here too, Amazon provides the lowest price for on-demand instances. Google Cloud offers the most aggressive discount for reserved instances, while Azure offers the lowest price for spot instances. 

Comparing On-Premises vs. Cloud Server Costs 

On-premises infrastructure refers to hardware and software that is owned and operated by an organization and kept on the organization’s own premises. Cloud infrastructure refers to hardware and software that is owned and operated by a third-party provider and made available to users over the internet.

There are several factors to consider when comparing the costs of on-premises versus cloud infrastructure:

Upfront Costs

On-premises infrastructure typically requires a larger upfront investment, as you need to purchase and set up the hardware and software yourself. This includes the cost of purchasing servers, storage systems, and other hardware, as well as the cost of installing and configuring the necessary software. In contrast, cloud infrastructure is usually paid for on a pay-as-you-go basis, which can be more cost-effective for organizations that do not have a lot of upfront capital.

New call-to-action

Maintenance and Support Costs

With on-premises infrastructure, you are responsible for maintaining and supporting the hardware and software yourself, which can be time-consuming and costly. This includes tasks such as patching and updating the software, repairing or replacing hardware that fails, and providing technical support to users. With cloud infrastructure, the provider typically handles maintenance and support, which can be more cost-effective.

Scalability

It can be expensive to scale up on-premises infrastructure, as you need to purchase and set up additional hardware and software as your needs grow. This can include the cost of purchasing new servers and storage systems, as well as the cost of installing and configuring the necessary software. In contrast, cloud infrastructure is generally more scalable, as you can easily increase or decrease your usage as needed, and you only pay for the resources you use.

Flexibility

With on-premises infrastructure, you are limited to the hardware and software that you have purchased and installed on your premises. If you need to add new capabilities or scale up your infrastructure, you may need to purchase additional hardware and software. In contrast, cloud infrastructure is generally more flexible, as you can easily access a wide range of resources and capabilities on demand.

The cost of on-premises versus cloud infrastructure will depend on your specific needs and requirements. It is important to carefully consider these factors and do a cost-benefit analysis to determine which option is best for your organization.

Related content: Read our guide to cloud cost optimization

Looking to Improve Your Application Performance?

LET’S CHAT
Scroll to top
Skip to content