Cloud Cost Optimization Rules
Continuous Monitoring to Reduce Your Cloud Bill
Continually monitor your cloud for cost reduction opportunities:
- Hyperglance is shipped with customizable rules, tailored to AWS & Azure
- New rules are added regularly in Hyperglance updates
- Trigger notifications and automate fixes using AWS SNS or Azure Event Grid notifications
AWS Rules
Application Load Balancer
Application Load Balancers with no Targets
Checks for any Application Load Balancers with no Targets.
Idle Application Load Balancers
Checks for any idle Application Load Balancers.
Aurora DB Cluster
Idle Amazon Aurora Clusters
Checks for any idle Amazon Aurora Clusters.
Classic Load Balancer
Classic Load Balancers with zero Instances attached
Unused Classic Load Balancers with no instances can be deleted to optimise your AWS expenditure.
Idle Classic Load Balancers
Unused resources should be deleted to save cost
DynamoDB Table
Unused DynamoDB Tables
Empty DynamoDB tables can still incur costs. Delete them to optimise your AWS expenditure.
EBS Snapshot
Orphan EBS Snapshots that are over 30 days old
Snapshots that are not associated with any current EBS Volume. Delete them to optimise your AWS expenditure.
EBS Volume
EBS Volumes Attached To Stopped EC2 Instances
Checks for any EBS Volumes Attached To Stopped EC2 Instances.
EBS Volumes Not Marked For Delete On Terminate
If an Amazon EBS volume isn't deleted when the instance that it's attached to is terminated, it may violate the concept of least functionality.
• CIS
• NIST 800-53
• NIST 800-171
• FedRAMP
Idle EBS volumes
Check to see if the Instance these EBS volumes are connected to is actually needed
IO EBS volumes
Check to see if the Instance these EBS volumes are connected to actually need these more expensive volumes.
Unattached EBS Volumes
If an Amazon EBS volume isn't deleted when the instance that it's attached to is terminated, it may violate the concept of least functionality.
• CIS
• NIST 800-53
• NIST 800-171
• FedRAMP
EC2 Instance
Idle EC2 Instances
EC2 instances that have used less than 10% on average over the past week.
Stopped EC2 Instances
Checks for any stopped EC2 Instances.
Elastic IP Address
Unattached Elastic IP Addresses
Checks whether all EIP addresses allocated to a VPC are attached to EC2 instances or are in-use.
• CIS
• NIST 800-171
• PCI DSS
NAT Gateway
Idle NAT Gateways
Checks for any idle NAT Gateways.
Network Load Balancer
Idle Network Load Balancers
Checks for any idle Network Load Balancers.
Network Load Balancers with no Targets
Checks for any Network Load Balancers with no Targets.
RDS DB Instance
Idle RDS Instances
Checks for any idle RDS Instances.
Previous Generation RDS Instances
Checks for any previous Generation RDS Instances.
RDS Instances that are Using Provisioned IOPS SSDs
Privisioned IOPS SSDs are only recommended for workloads requiring 10000 IOPS or throughputs of 160MiB/s.
Redshift Cluster Node
Idle Redshift Cluster Nodes
Checks for any idle Redshift Cluster Nodes.
Underutilized Redshift Cluster Nodes
Checks for any underutilized Redshift Cluster Nodes.
Workspace
Unhealthy Workspaces
Checks for any unhealthy Workspaces.
Workspaces with no user activity in the last 14 days
Checks for any Workspaces with no user activity in the last 14 days.
Azure Rules
Application Gateway
Idle Application Gateways
Cosmos DB Account
CosmosDB not being used
Underutilized CosmosDB
Disk
Unattached Disks
Load Balancer
Idle Load Balancers
MySQL Server
Underutilized MySQL Servers
PostgreSQL Server
Underutilized PostgreSQL Servers
Public IP Address
Unassociated Public IP Addresses
Delete Public IPs that are not being used to save money
SQL Server
Underutilized SQL Servers
Traffic Manager Profile
Idle Traffic Manager Profiles
Virtual Machine
Idle Virtual Machines
Stopped Virtual Machines
Web App
Idle Web Apps
metric("Requests", Average, Week) < 10.0
