Scribd
Upload
115 views

Module6 HA PDF

This document discusses various techniques for achieving high availability on AWS, including using reserved instances, elastic IP addresses, elastic load balancers, auto-scaling groups, multi-AZ RDS deployments, and example reference architectures. It provides links to documentation and demos of setting up high availability for single EC2 instances, across availability zones, within a single AZ, across multiple AZs, and using auto-scaling groups across AZs with a multi-AZ RDS backend. Assignments are given to implement the scenarios hands-on and do a cost comparison.

Uploaded by

srinubasani
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
115 views

Module6 HA PDF

This document discusses various techniques for achieving high availability on AWS, including using reserved instances, elastic IP addresses, elastic load balancers, auto-scaling groups, multi-AZ RDS deployments, and example reference architectures. It provides links to documentation and demos of setting up high availability for single EC2 instances, across availability zones, within a single AZ, across multiple AZs, and using auto-scaling groups across AZs with a multi-AZ RDS backend. Assignments are given to implement the scenarios hands-on and do a cost comparison.

Uploaded by

srinubasani
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 7

Module 6: High Availability 

Introduction 
Reserved Instances 
https://aws.amazon.com/ec2/purchasing­options/reserved­instances/ 
 
Elastic IP address 
http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/elastic­ip­addresses­eip.html 
 
Elastic Load Balancer  
https://aws.amazon.com/elasticloadbalancing/ 
 
Auto­Scaling group 
https://aws.amazon.com/autoscaling/ 
 
RDS Multi­AZ deployment 
https://aws.amazon.com/rds/details/multi­az/ 
 
Reference high availability architecture 
http://media.amazonwebservices.com/architecturecenter/AWS_ac_ra_ftha_04.pdf 
 
CLI 
Auto­scaling: ​
http://docs.aws.amazon.com/cli/latest/reference/autoscaling/index.html 
Elastic load balancer: ​
http://docs.aws.amazon.com/cli/latest/reference/elb/index.html 

   
Demo 

Recover from hardware failure 
Single EC2 instance with Elastic IP address. Create replacement EC2 instance and attach EBS volume from 
old instance, associate Elastic IP to new EC2 instance. 
 

 
 
Preparatory Steps: 
● Launch EC2 instance in an AZ 
● Allocate new Elastic IP address 
● Associate Elastic IP address with EC2 instance 
Failover Steps: 
● Stop primary EC2 instance 
● Detach it’s volume 
● Terminate primary EC2 instance 
● Launch new EC2 instance in same AZ 
● Stop new EC2 instance (to replace EBS volume) 
● Detach volume from new EC2 instance 
● Attach volume of primary EC2 instance to new EC2 instance 
● Start new EC2 instance 
● Associate Elastic IP address with new EC2 instance 
Recover from AZ failure 
EC2, RDS and Elastic IP in a single AZ. Create standby EC2 instance in another AZ and keep it in stopped 
state. Upon failure in primary AZ, create RDS instance in another availability zone using RDS snapshot. 
Associate Elastic IP address to standby EC2 instance after starting the instance.  
 
 

 
 
Preparatory Steps: 
● Create AMI from primary EC2 
● Create snapshot from primary RDS 
● Launch standby EC2 instance using AMI in standby AZ 
● Keep standby EC2 instance in stop state (ready for use if primary fails) 
Steps on AZ failure: 
● Start standby EC2 instance 
● Launch RDS instance in standby AZ using primary RDS instance’s snapshot 
● Dissociate Elastic IP from primary EC2 and associate to standby EC2 
● Terminate EC2 and RDS in primary AZ   
High availability in single AZ 
Create two EC2 instances behind an ELB with single AZ RDS. 
 

 
 
Steps 
● Launch 2 EC2 instances using your AMI in a single AZ 
● Launch RDS instance in same AZ 
● Launch ELB and associate with 2 EC2 instances 
 

   
Multi­AZ EC2 and RDS 
Create two EC2 instances behind an ELB, one instance each in two AZs, and with Multi­AZ RDS. 
 

 
 
Steps: 
● Launch one EC2 instance each in two AZ 
● Launch multi­AZ RDS 
● Create cross­AZ ELB and associate with 2 EC2 instances 
 
 

   
Multi­AZ with Auto­Scaling Group 
Create an auto­scaling group with minimum 2 and maximum 8 EC2 instances behind an ELB. Instances 
should be evenly deployed in two AZs. Create multi­AZ RDS instance. 
 

 
 
Steps 
● Create VPC with 2 subnets in 2 AZs 
● Launch multi­AZ RDS in VPC, primary in one subnet and secondary replica in second subnet 
● Launch cross­AZ ELB 
● Create multi­AZ auto­scaling group in the VPC with min 2 instance and max 8 instances. Auto­scale 
based on CPU utilization thresholds 
● Associate ELB with Auto­scaling group 

   
Assignments 
● Repeat above scenarios using your own AWS account. 
● Do cost comparison between all scenarios listed above. 
● Use Route 53 to create a multi­region high availability architecture. Hint: Use 2 Elastic load balancers, 
running EC2 instances in multiple AZs within each Region. 
 

You might also like