Lecture # 17 : Cost and Capacity Management for Compute Services
cloud PATH-AWS

5 minutes


go back go back go back home home

Introduction

Cost Management tackles the question of “how to save money?”, while Capacity Management answers “How do we meet the demand of traffic & usage through adding/upgrading services?“.

EC2 Spot Instances / Reserved Instances / Savings Plan

One way to save on compute services is by paying up in full or partially(Saving Plan), or by commiting to yearly contracts (RI), or by being flexible about availablity and interruptions to compute services (spot instances).


AWS Batch

Plan schedules and execute your batch computing workloads across full range of AWS compute services, can utilize spot instances to save money.


AWS Compute Optimizer

It is a service which goes through your historic usage and suggests how to reduce costs and improve peformance, by using ML to analyze your previous usage history.


AWS Auto-Scaling-Group

Automatically adds/removes EC2 servers to meet the current demand of traffic. Will save you money and meet the capacity since you only run the amount of servers you actually require.


AWS Elastic Load Balancer

Distribute traffic to multiple instances, can reroute traffic from unhealthy instance to healthy instance. Can route traffic to EC2 instance running in different AvailablityZones.


AWS Elastic Beanstalk

It is for easily deploying web applications without developers having to worry about understanding the underlaying AWS service. Similar to Heroku.