AWS Pricing concepts

                                AWS Pricing concepts

Pricing

Amazon Web Services offers a broad set of global compute, storage, database, analytics, application, and deployment services that help organizations move faster, lower IT costs, and scale applications.

Pricing for all services is unique and there are not complex dependencies or licensing requirements. Estimation of Pricing is one of the crucial concepts in AWS. The AWS Simple Monthly Calculator is used for this purpose to estimate your monthly bill. It provides a per-service cost breakdown, as well as an aggregate monthly estimate. You can also use the calculator to see an estimate and a breakdown of costs for common solutions. This particular field walks you through an example of using the AWS Simple Monthly Calculator to estimate your monthly bill.

AWS offers three different concepts of Pricing. They are :

• On-demand Instances
• Reserved Instances
• Spot Instances

Note: AWS pricing you see in this documentation, was related to at the time of publication. For the latest pricing information, go to AWS Service Pricing Overview. For more information about AWS pricing works, go to How AWS Pricing Works.

Amazon EC2 Cost Breakdown

The following table shows the characteristics for Amazon EC2 which we have identified for this web application architecture. In this example, we’ll assume that you’ve moved into full production and you need between three and six instances. Three instances run all the time, two additional instances are required to handle peak traffic times, and another instance handles nightly backups.

Characteristic	Metric	Description
Clock Hours of Server Time	3 instances running 24 hours/ day 2 instances running 8 hours/ day 1 instances running 3 hours/ day 1 instances running 3 hours/day	Assuming an average of 30.5 days in a month, the full-time instances run 732 hours/ month, the peak traffic instances run 244 hours/ month, and the backup instances run 91.5 hours/ month
Machine Characteristics	1 ti.micro instance 5 m1. small instances	Micro – 613 MB of memory, up to 2 ECUs (for short periodic bursts), EBS storage only, 32-bit or 64-bit platform Small- 1.7 GB of memory, 1 EC2 Compute Unit (1 virtual core with 1 EC2 Compute Unit), 160 GB of local instance storage, 32-bit platform
Additional Storage	1 EBS Volume Storage: 10 GB/ Month 100 IOPS	The AMI is EBS-backed. The volume will have 10 GB provisioned storage, and 100 I/O requests per second made to the volume.
Data Transfer	Data In: 0.005 GB/ day Data Out: 0.05 GB/ day	There are approximately 1,000 hits per day. Each response is about 50 KB, and each request is about 5 KB.
Instance Scale	Between 3 and 6 instances	You need 3 instances running all the time, another two to handle peak traffic, and another one to handle nightly backups.
Elastic Load Balancing	Hourly usage: 732 hrs/ month Data processed: 1.525 GB/ month	Elastic Load Balancing is used 24 hrs/ day, 7 days/ week Elastic Load Balancing processes a total of 0.055 GB/ day (data in + data out)

The following image from the AWS Simple Monthly Calculator shows the cost breakdown for Amazon EC2.

The total monthly cost is the sum of the cost of the running instances, EBS volumes and I/O requests, Elastic Load Balancer and the data processed by the Elastic Load Balancers. Because we used basic monitoring and only one metric and alarm for our Amazon EC2 instances, there is no additional charge for Amazon CloudWatch monitoring.

Variable	Formula	Calculation
Instance Cost	Instance cost per hour Number of instances x Clock hours of server time ————————	$ 0.0803 x 732 ——- $ 175.68
Instance Cost	Instance cost per hour Number of instances x Clock hours of server time ————————	$ 0.0802 x 244 ——- $ 39.04
Instance Cost	Instance cost per hour Number of instances x Clock hours of server time ————————	$ 0.021 x 91.5 ——- $ 1.83
Additional Storage	Storage rate x Storage Amount (GB)+ (I/O requests rate x seconds per month x Request rate (per 1M requests)) x Number of Volumes ——————————————————–	$ 0.10 X 10+ (100 x 2.6M x $ 0.10)/ 1M x 6 ———————– $ 162.00
Elastic Load Balancing	Hours used x Hourly rate + (Data processed (GB) x Process rate) ————————————–	732 x $ 0.025+ 1.6775 x $ 0.008 —————– $ 18.31
Total Cost Per Month		$ 396.86

Moving on to Summing up, view a summary of the total charges, including AWS Data Transfer out and the free usage tier discounts.

Summing It All Up

The SUM function returns the sum of the input column or expression values. The SUM function works with numeric values and ignores NULL values.

To calculate the total cost for this example, we add the cost for Amazon EC2 instances and the AWS Data Transfer Out and then subtract any discount that falls into the AWS free usage tier. To learn more about the free usage tier and to find out if you are eligible, go to ‘Getting Started Guide AWS Free Usage Tier‘.

The total AWS Transfer Out is an aggregate Data Transfer Out usage across all Amazon EC2 instances. For Amazon EC2, we have 0.05 GB per day, which is approximately 1.525 GB per month. Because up to one month 1 GB of data transferred out is free, then we would be left with a total of 0.525 GB per month.

Variable	Formula	Calculation
AWS Data Transfer	(Data in (GB) X Data In Rate) + (Data out (GB) X Data Out Rate) ———————————	0.1525 X $ 0.00+ (0.525) X $ 0.12 —————– $ 0.06

The following image shows an example of your monthly estimate.

According to the calculator, the total cost for Amazon EC2 is summed up as $373.50.

How To Further Save Costs

In the example deployment, which we have been discussing, we used On-Demand Instances for all our six instances. With On-Demand Instances, you are charged only from the time you launch an instance, until the time you terminate it. If you are planning to run your instances for a long time, you can save more money by reserving them.

To obtain Reserved Instances, you make a low, one-time payment for each instance, which you want to reserve. In return, you will receive a significant discount on the hourly usage charge. If you know approximately how heavily your Amazon EC2 instances will be used when they are running, you can save even more by opting for Heavy, Medium or Light Utilization Reserved Instances.

With Heavy Utilization, you pay a higher upfront fee, but normally your hourly usage rate is lower than that of Medium and Light Utilization Reserved Instances.

Light Utilization has the lowest upfront fee, but your hourly rate is higher than that for Medium and Heavy Utilization Instances. In the previous example, three of the instances are running all the time. These instances are ideal candidates for Heavy Utilization Reserved Instances. Two instances run only during peak traffic, about one third of the time. They are ideal candidates for Light Utilization Reserved Instances. Because the instance that performs the nightly backups runs only a few hours a day, you can run it as an On-Demand Instance.

Reserved Instances can be obtained in 1-year or 3-year terms. The 3-year term can offer additional savings, over the 1-year term. For more information about reserving instances, go to Amazon EC2 Reserved Instances. You can see the cost comparison with On-Demand versus Reserved Instances over a three-year period in the following table.

Using the same characteristics and metrics in the above example, let’s update the calculator to enter the Heavy and Light Utilization as in the following diagram.

The total monthly cost is calculated the same way as the previous example, except that there is an additional one-time fee for Reserved Instances. The total cost is as shown in the following diagram.

The following table compares the total costs for using a mix of Heavy and Light Utilization Reserved Instances with those for On-Demand Instances.

Instance	Monthly Cost	One-time Fee	Total Cost (3 years)
6 On-Demand instances	$ 373.50	n/ a	$ 13446.00
1 On-Demand Instance: 3 Heavy Utilization Reserved Instances 2 Light Utilization Reserved Instances	$ 202.46	$ 1112.60	$ 8401.16

As you can see from the table, by using a mix of Heavy and Light Utilization Reserved Instances in this example, you can save approximately 30%. For more information on how AWS pricing works, go to the How AWS Pricing Works white paper.

Another way you can save money is by using Spot Instances. Spot Instances are unused Amazon EC2 capacity that you bid for. Instances are charged at the Spot Price, which is set by Amazon EC2 and fluctuates periodically depending on the supply of, and demand for, Spot Instance capacity. If your maximum bid exceeds the current Spot Price, your bid request is fulfilled, and your instances will run until either you choose to terminate them or the Spot price increases above your maximum bid, whichever is sooner. To learn more about spot instances, go to http://aws.amazon.com/ec2/spot-instances.