1. Daily Inventory of all AWS resources:
   • EC2 instance, including modelling to rightsize instance
   • RDS instances, including modelling to rightsize instance
   • Elasticache, including modelling to rightsize instance
   • EBS volumes
   • Elastic Load Balancers
   • S3
   • VPCs
   • Subnet
   • Routing Tables
   • IAM
   • Heatmap - for each hour number of instances running for each instance type
   • Heatmap.cost - for each hour cost for all instances for each instance type
   • Performance utilisation of above resources based on 95th percentile, proposed rightsizing of instances
   • Cost for each resource and total cost for each AWS service
   • When instances running (day/night), percentage
2. Performance utilisation of above resources based on 95th percentile
3. Cost for each resource and total cost for each AWS service
4. When instances running (day/night), percentage

Why?

• CMDB of AWS resources
• Right size resources based on historical performance data and save costs
• Using heatmap bather baseline usage of running instances to workout right amount of Reserved Instances
• Stop instances at night/weekends
• Terminate/stop unused instance/volumes/snapshots

How to get it running:

• Install AWS CLI
• create AWS CLI profiles for all required accounts
• It uses python library xlwt to create Excel spreadsheet. To install: pip install xlwt
• Customise three environment variables in awsinfo.sh: - basedir: where awsinfor will reside - accounts: list of AWS accounts (or AWS CLI profiles) - s3bucket: name of s3bucket where to upload Excel spreadsheet
• customise aws.costs (listed file contain Sydney region costs)
• Add awsinfo to daily cron job such as: 0 11 * * * /usr/local/awsinfo/awsinfo.sh > /usr/local/awsinfo/awsinfo.out 2>&1