The bill from AWS is similar to an electricity bill. The services are billed based on usage. You pay based on how long the virtual machine is running, the amount of storage space, or the number of load balancers. Services are billed monthly. Pricing per service is always public, if you want to calculate the monthly cost of a predefined setup you can use the AWS Simple Monthly Calculator .
Free Tier (free tier)
You can use AWS service free of charge for 12 months from the date of registration. The idea behind Free Tier is to allow you to experiment with AWS and get some experience using that service. Here are the services Free Tier offers:
- 750 hours (about 1 month) for a small virtual machine running on Linux or Windows. This means 1 virtual machine running continuously for 1 month or 750 virtual machines running for 1 hour.
- 750 hours (about 1 month) for an application load balancer.
- Object store 5GB.
- Small database with 20GB of storage, including backup.
If you go over the Free Tier limit, you will start paying for resources usage without notice. You will receive an invoice at the end of the month. Before using AWS, you will be taught how to track expenses.
After your 1 year trial period ends, you pay for all the resources you use. But some resources are permanently free. For example, 25 GB NoSQL database.
You can find out more about Free Tier at Free Tier . You will be notified when you use the parts that do not support Free Tier, so don’t worry about billed parts.
The fees you have to pay
As mentioned, you may be charged in the following cases:
- Based on minutes or hours used: Virtual machines are billed per minute. Load balancers are billed by the hour
- Traffic Based: Traffic is calculated by Gigabytes or by the number of requests (requests).
- Based on memory usage: Usage can be measured in terms of capacity (eg you are given 50GB of space and it doesn’t matter how much 50GB you use). Or can be calculated according to actual usage (for example, if you use 2.3GB, it is only over 2.3GB).
We will base on the previous web shop example and add the fee as follows:
Let’s say that the web shop started running in January, and you have a marketing campaign to increase sales for the next month. How lucky! You can increase your traffic fivefold by February. As you know, AWS charges usage based fees.
Based on the fee table we see: the number of visits increased from 100,000 to 500,000, and your monthly bill increased from $ 127 to $ 495 USD (up 3.9 times). Because the web shop handles more traffic, you pay extra for services, such as CDNs, web servers, and databases.
Other services like storage for static files have not increased, so the price remains the same.
With AWS, you can achieve a linear relationship between traffic and costs. And another opportunity awaits you with this pricing model.
Usage-based pay advantage
The AWS-based pricing model creates new opportunities.
The barrier to starting a new project is reduced, for example, as you are not required to invest in infrastructure before starting the project. You can initialize the virtual machine on demand and pay per second of use, and you can turn off the virtual machine at any time to stop charging. You do not need to pre-commit your memory usage.
Another example. A large server costs and capacity equal to 2 small servers. So you can break the system down into smaller chunks, because the costs are the same. This makes your system fault tolerant. With this advantage your company does not need to be a large-scale company but a small company can still take advantage of AWS’s advantages.
AWS is not the only cloud service provider. Microsoft Azure and Google Cloud Platform (CGP) are also major providers.
The 3 cloud providers have the following in common:
- The global infrastructure provides computing, networking, and storage capabilities.
- Infrastructure as a Service IssS (Infrastructure as a Service) provides virtual machines on demand: Amazon EC2, Azure Virtual Machines, Google Compute Engine.
- Unlimited storage scalable, highly decentralized storage system: Amazon S3, Azure Blob storage, Google Cloud Storage.
- Usage-based charging model. (like charging for electricity – how much to use, how much you can use a subscription for 1 year – 3 years to get a discount)
But what makes these cloud providers different?
AWS is the market leader in cloud computing, Providing an extensive product portfolio. Even as AWS has expanded into the enterprise space lately, it’s clear AWS is starting with its internet-scale problem solving service. In short, AWS is building great service based on cutting edge technologies, mostly open source. AWS provides a complex service way but in a solid and secure way to restrict access to your cloud.
Microsoft is providing Microsoft’s technology platform to the cloud, recently focusing on open source technology and the web. It seems that Microsoft is working hard to catch up with AWS’s cloud computing market share.
GCP is focused on developers who want to build complex distributed systems. Google incorporates their global infrastructure to provide a scalable and error-tolerant service (such as Google’s load balancing service). GCP is more focused on cloud-native apps than moving your local server to the cloud.
There are no shortcuts to making an informed choice about which supplier to choose. Because each case and project is different. The problem is in the details. And don’t forget which platform you are using. (Are you using a lot of Microsoft technology? You have a large team of system administrators or you are a developer focused company). In short, AWS is the most mature and powerful cloud platform available today.
Thank you for watching.
… To be continued …
Thank you for your interest. If you find it interesting, please leave an email below, I will email you when the next section is complete.
Reference source: Amazon Web Services in Action, 2nd Edition (Michael Wittig and Andreas Wittig).