Auto Scaling Group(ASG)

-

 Auto Scaling Group(ASG)




What is Auto Scaling ?
  • Auto Scaling monitor your application and adjusts capacity automatically to ensure consistent, predictable performance at the lowest possible cost
  • It is simple to set up application scaling for multiple resources across multiple services using AWS autoscaling in minutes
  • Scaling is he process of adding and removing resources as needed
  • Scale out is removing resources and scale in is adding resources
  • Types - Vertical and Horizontal


Vertical Scaling - 
  • We having one single machine and we just increase the capacity of machine by adding more resources likes CPU, Storage etc.

Horizontal Scaling - 
  • We are adding more number of machines or nodes at the same time to existing pool of resources



Lifecycle of Auto Scaling -  



Auto Scaling Components - 

Auto Scaling Groups - 
  • Auto scaling group contains the collection of EC2 instances that are exactly the same 
  • While creating ASG launch configuration must be specified, after specifying the launch configuration cannot be change 
  • New instances are launched using a new configuration 
  • EC2 instances are launched and terminated using scaling policies


Configuration Template - 
  • Launch configuration is a template that is used to to launch EC2 instances for the autoscaling purpose 
  • Launch configuration cannot be modified after creation
  • It can be created in two ways 
  1. From scratch - Image type, instance id, storage etc.
  2. From EC2 Instance - Attributes from the inctance are copied. Block device mapping of AMI is included. Any additional device that are attached after launching the instances is not considered in the launch configuration


Scaling Policies - 

Scheduled Scaling - 
  • Scale based on a schedule
  • Used when the load is predictable
Simple Scaling - 
  • Scale to certain size on a cloudwatch alarm
  • Ex. when CPU > 90%,  then scale to 10 instances
Step Scaling - 
  • Scaling in steps using cloudwatch alarms
  • Specify the instance warmup to scale faster
  • Ex. when CPU 70% >,then add 2 units and when CPU <30%,then remove 1 unit
Traget Tracking Scaling - 
  • ASG maintains the cloudwatch metrics and scale accordingly (automatically creates cloudwatch alarms)
  • Ex. maintain CPU usage usage at 40%
Predictive Scaling - 
  • Historical data is used to predict the load using ML and scaling automatically



Scaling Cooldown - 
  • After a scaling happens, the ASG goes into cooldown period (default 300 sec) during which it does not launch or terminate additional instances (ignore scaling request) to allow the metrics stabilize
  • Use the ready-to-use AMI to launch instances faster to reduce the cooldown period


Health Checks - 
  • By default ASG uses EC2 status check (not the ELB health check). This could explain why some instances that are labelled as unhealthy instances by an ELB are still not terminated by ASG 
  • To prevent ASG from replacing unhealthy instances, suspend the ReplaceUnhealthy process type


Instance Termination ( Termination Policy) - 
  • Select the AZ with most number of instances
  • Delete the instance with the oldest launch configuration 
  • Delete the instance which is closest to the next billing hour
  • Flow diagram




Rebalancing AZs - 
  • ASG ensures that the group never goes below the minimum scale. Actions as changing the AZ for the group or explicity terminating or detaching instances can lead to the ASG becoming unbalanced between AZs. In such cases, ASG compensates by rebalancing the AZs by launching new instance before terminating the old one, so that rebalancing does not compromise performance or availability of the application 


Attach running instances to ann existing ASG -

The running instance must meet the following criteria - 
  • The AMI used to launch the instance still exists
  • The instance is not the member of another ASG 
  • The instance is launched into one of the AZ defined in your ASG
 



Comments

Post a Comment

Popular posts from this blog

AWS Instance Store

-
Image
 AWS Instance Store  AWS Instance Store -  Hardware storage directly attached to EC2 instance amd cannot be detached and attached to another instance  Highest IOPS for any available storage Ephemiral Storage (Loses data when instance is terminated, stopped or hibernated) Good to buffer/cache/scratch data/temporary content  AMI created from an instance does not its instance store volume preserved  You can specify the instance store volumes to an instance when you launch the instance. You can't attach instance store volumes to an instance after you’ve launched it. Raid -  Raid 0  Improve performance of a storage volume by distributing reads and write in a stripes across attached volumes  If you add a storage volume you get the straight addition of throughput and iops For high performance applications Raid 1   Improve data availability by mirroring in multiple volumes For critical applications
Read more

AWS Identity and Access Management

-
Image
  IAM Introduction to IAM -  AWS identity and management is a web service that helps you to securely control access to AWS resources You use IAM to control who is authenticated and authorized to use resources IAM allows you to create and manage AWS users and groups , and use permissions to allow or deny their access to AWS resources IAM is a fundamental part of AWS security, as it allows you to implement the principle of least privilege , ensuring that users and applications have only the permissions they need to perform their tasks IAM also provides features such as multi-factor authentication (MFA) , password policies, and access keys to enhance the security of your AWS environment Features of IAM -  IAM User  -   IAM allows you to create and manage users in your AWS account Each IAM user has a unique name and security credentials, which are used to authenticate and authorize their access to AWS resources IAM users have security credentials that are us...
Read more

Elastic Block Storage (EBS)

-
Image
 Elastic Block Storage (EBS) Elastic Block Storage (EBS) -  AWS Elastic Block Storage is a web service that provides block level storage volumes for use with amazon EC2 instances Amazon EBS volumes are highly available and reliable storage volumes that can be attached to any running instance and used like hard drive Amazon EBS allows you to create storage volumes and attach them to EC2 instances   It is used for various databses (both relational and non relational) as well as a wide range of application such as software testing and development  EBS provides a high performance option for many use cases it is directly attached to the instance Features of EBS  -  Amazon Elastic Block Storage  Region and availability zones are multiple physical location for your resources such as instances and EBS volumes Using security groups, you can specify the protocols, ports and source IP ranges that can reach your instances Elastic IP addresses are static IPv4 ...
Read more