Out of the big 3 cloud platform providers, Amazon Web Services (AWS) was the first to render cloud services, with the Google Cloud Platform (GCP) and Microsoft Azure following thereafter. Overtime, each of the three cloud platforms have become well-established and continue to introduce new and innovative services to help organizations achieve their business goals. As each platform becomes more popular, we need to better understand which platform offers the most competitive pricing. We are going to take a look at how each platform ranks in terms of their costing for a simple solution, to determine which cloud provider is the most cost-effective.
Case study:
A solutions architect needs to design a cloud-based architecture for a website hosted on the cloud.
The website is required to perform the following services:
- Login/Register a user
- Update a user’s personal information
- View user details
- Upload documents
The solution architect wants to avoid designing a monolithic application to reduce the number of potential failover areas. The solutions architect wants to ensure that user details are stored in a database. The solutions architect would like to determine which of the big three cloud platform providers should be used for the most cost-effective solution. The online store is believed to become very popular, because of the unpredictable demand the solution architect would like to ensure that the services used are scalable.
Let’s first take a look at a GCP-based solution.
Architecture Interpretation:
- The user will send a request to GCP’s Cloud Load Balancing service when accessing the website, which allows you to balance traffic to various App Engine instances
- The website’s microservices are containerized and built as a Docker image
- The Docker image is then pushed into the Container Registry
- Cloud Build then pulls the image from the Container Registry and builds it
- The image is then deployed to App Engine
- User details are stored in CloudSQL
Cost of Services:
Service | Assumptions | Cost | Total Cost per Month (30 days) |
Cloud Load Balancing | Iowa region (us-central-1) 10 forwarding rules with 2,048 GiB of network ingress | – | USD 54.77 |
Container Registry | 4 x 10GB containers | USD 0.026 per GB per month | USD 10.40 |
App Engine | 2 x F1 instances per hour | – | USD 30.44 |
Cloud Build | 1 x vCPU 4GB Memory 1000 build-minutes per day | USD $0.003 / build-minute (First 120 build-minutes are free) | USD 79.2 |
CloudSQL | 4GB Memory 1 vCPU | USD 5.11 per GB USD 30.15 per vCPU per month | USD 50.59 |
USD 225.40 |
The total for this project using GCP services is USD 225.40 per month. Next we will take a look at the cost associated with using AWS and Azure, for the same project with the same specifications. By determining the pricing of each platform for a simple cloud-hosted website, we can get a better understanding of their pricing structures and an idea of which of the three platforms is, in essence, the most affordable.
