Now that we have come to the end of this chapter, let us look at some key recommendations and best practices that you need to keep in mind when working with ELBs and Auto Scaling:
- Plan and provide the ELB with enough of a grace period (by default its 300 seconds) so that it does not put an instance in the unhealthy state even before the application has had time to initialize completely.
- Use Amazon Route 53 and provide a suitable domain name for your applications. Additionally, leverage Route 53 to balance your application's load across multiple regions as well.
- Although ELBs can handle large loads (up to 20k/sec), they can only do so if the load increases gradually, say over a period of several hours. If your application spikes in load in minutes rather than hours, then you are better off by using pre-warmed ELBs. To know more about pre-warmed ELBs and how to get them, refer to http://aws.amazon.com/articles/1636185810492479#pre-warming.
- Configure HTTPS and SSL listeners for your ELB whenever possible.
- Plan your Auto Scaling well in advance. This includes deciding on the number of instances that will be required for your application as well as the type of instance family.
- Plan on which monitoring metrics (CPUUtilization, MemoryUtilization, Disk Space used, and so on) you tend to use and set up the scaling policies accordingly.
- Deploy your Auto Scaling Groups across multiple AZs. This provides an additional layer of high availability in case an entire AZ should fail.
- Prepare, test, and bootstrap your application on an AMI before adding it to the Auto Scaling activity. Try and keep your application as decoupled as possible.
- Always monitor and set up notifications for your Auto Scaling activity. This will help you track and maintain your application's as well as instances' performance.
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.