Add & Configure Services

Workload Description

Lets take the case of a customer facing Web Application. This general purpose workload takes input data from users (over the internet), processes it and returns the results. It is a spiky workload which receives 100 new connections per second, each lasting approximately 3 min. Per connection, the workload processes 1000 bytes of data across 4 requests per sec. The workload requires 2 instances at peak, with 2 GB RAM, 2 vCPU each and 30 GB of storage per instance. The workload needs a 100 GB database which can support transactional traffic.

A simple 3 tier LAMP (Linux Apache MySQL PHP) stack based Web Application on AWS uses Amazon Application Load balancer, Amazon EC2, Amazon RDS MySQL (Relational Database Service). We will now add and configure these 3 services in the Pricing Calculator.

Add & Configure Services

Add Load Balancer

1. On the Add Service page, choose Elastic Load Balancing by clicking Configure on that tile. You can also use the search bar by typing Load Balancer to narrow down the results.

3. For the Description , enter “Load Balancer”

4. Choose US West (Oregon) for the Region

5. In the Elastic Load Balancing section, choose Application Load Balancer. Also choose Load Balancer in AWS Region

6. In the Service settings section, enter “1” for Number of Application Load Balancers

7. In the Load Balancer Capacity Units (LCUs) section:

• Skip Processed bytes (Lambda functions as targets) since we are using EC2 instances
• For Processed bytes (EC2 Instances and IP addresses as targets) enter “0.36” GB per hour . We get this number since 1,000 bytes of data is processed connection.
• For Average number of new connections per ALB enter “100” connections per second
• For Average connection duration enter “3” minute
• For Average number of requests per second per ALB enter “400”
• Rules determine how the load balancer routes requests. For example, the default rule only routes HTTP traffic on port 80 to the EC2 instances (targets). Enter “20” for Average number of rule evaluations per request

8. Click on Add to my estimate

Add EC2

1. On the My Estimate page, click on Add Service.

2. Choose EC2 by clicking Configure on that tile. You can also use the search bar by typing EC2 to narrow down the results.

3. For the Description , enter “EC2”. Leave the default value for Region to be US West (Oregon)

4. Click on Advanced estimate :

5. In the EC2 instance specifications section, choose Linux for the Operating system

6. In the Workload section, you can select the pattern that best describes the workload.

• Select Daily spike traffic
• Expand the section Daily spike pattern by clicking on the arrow
• Leave the workloads days to the default option - Monday to Friday
• Enter 1 for the Baseline and 2 for the Peak, indicating that this workload requires 1 instance at normal times and 2 instances during peak.
• Leave the default value - 8 hrs and 30 min for Duration of peak

7. In the EC2 Instances section, you can choose the instance needed for this workload. Given that this is a spiky workload, the t instance family is a good fit.

• Choose t4g.small , which has the requires 2 vCPU, 2 GB RAM

8. In the Pricing Strategy section, you can choose the option that best fits your need. For this example, we will choose On-Demand. Once deployed, you can use the Lab on Pricing Models to analyze and determine the best Pricing Strategy for this workload.

9. In the Amazon Elastic Block Storage (EBS) section, configure the storage required for this workload

• Choose gp3 for the storage type
• Leave the default values for IOPS and Throughput - which is sufficient for this workload.
• Enter 30 for Storage amount
• Lets chose a daily backup schedule. Choose Daily for Snapshot Frequency and enter “1” GB for Amount changed per snapshot

10. In the Data Transfer section, you can specify the networking requirements for this workload

• Enter 50 and GB per month for Inbound Data Transfer
• Enter 200 and GB per month for Outbound Data Transfer

11. Click on Add to my estimate

Add RDS

1. On the My Estimate page, click on Add Service.

2. Choose RDS for MySQL by clicking Configure on that tile. You can also use the search bar by typing RDS to narrow down the results.

3. For the Description , enter “Database”. Leave the default value for Region to be US West (Oregon)

4. Choose US West (Oregon) for the Region

5. In the section MySQL instance specifications:

• Enter 1 for Quantity
• For this workload, a m6g.large database instance would be sufficient. Choose db.m6g.large
• Choose Multi-AZ for Deployment Model
• Leave the default options for Pricing model, Term and Purchase option

6. In the section Storage:

• Choose General Purpose SSD (gp2) for Storage for each RDS instance
• Enter 100 GB for Storage amount

7. For this workload, the default RDS backups are sufficient. You can skip the Backup section.

8. Click on Add to my estimate