AWS Classroom Series – 26/Nov/2020

AWS Instance LifeCycle

• Instance States
• pending:
  • This is a transition state preparing to enter running state
  • You will not be billed in this state
• running:
  • The instance is ready & running
  • You will be billed in this state
• stopping:
  • This is a transition state to perpare for stopped or stop-hibernate
  • You will not be billed for stopped but billed for hibernation
• Stopped:
  • The instance is shutdown
  • No billing
• Shutting down:
  • Tranisition state for terminated
  • Not billed
• Terminated:
  • Instance is permanently removed
  • Not billed
• Note: This doesnot apply for reservations.

AWS Instance Purchasing Options

• AWS provides the following instance purchase options
  • On-Demand Instance:
    • Pay by second/hour of instance usage
  • Savings-Plan
  • Reserved Instances (RI):
    • In Reserved instance model, we give commitment to amazon on particular instance type for a period of 1 or 3 years
    • Payment can be made
      • Complete upfront
      • Partial Upfront
      • Pay as you go
    • Amazon gives the following flexibility in Reserved Instances
      • AZ flexibility: The RI instance discount applies to instance usage in any AZ in the specified region
      • Instance size flexibility: RI discount applies to instance usage within instance family regardless of size (Supported on Amazon linux/Unix instances)
  • Spot Instances:
    • We can reserve the unused hardware in AZ’s at very less cost
    • We can bid for a price of ec2 instance type with some price (usually very less)
    • when the ec2 instance type prices is equal to your bid price ec2 instance will be automatically created.
    • Once the price of ec2 instance increases (due to other bids), then ec2 instance will be evicted from your account.
  • Dedicated Hosts: EC2 dedicated host is a physical server with EC2 instance capacity fully dedicated to your use
  • Dedicated Instances: Dedicated instances are ec2 instances that run on hardware dedicated to single customer
  • Capacity Reservations

Understanding ec2 pricing

• Scenario-1:
  • My Team has to create test environment for the following architecture. You are asked to create an estimate of pricing

    • 2 Web Servers: 2 VCPUs & 8GB of RAM

    • 2 App Servers: 4 VCPUS & atleast 8 GB of RAM

    • 2 db Servers: 4 VCPUS and 16 GB of RAM

  • Since the environment is test, lets use general purpose compute
    • Webserver (t2.large => 2)
    • app server (t2.xlarge => 2)
    • db server (t2.xlarge => 2)
  • Lets create an estimate (excluding storage costs)
  • Navigate to pricing calculator Refer Here
  • On-demand estimate
• Scenario-2:
  • Create the estimate for the same resource as in scenario 1 for production env
    • Web Server: Compute optimized
    • App Server: Compute Optimized
    • Db Server: Storage Optimized
• Create estimates for Scenario 2 with
  • Savings plan
  • Reserved Instances partial upfront & Complete upfront and see the price differences.