Introduction
Managing resources efficiently is crucial in any cloud environment. In the context of AWS WorkSpaces, keeping track of CPU utilization and making timely tier changes can help in optimizing costs and performance. This article presents a CloudFormation template that automatically manages AWS WorkSpaces tiers based on CPU utilization.
AWSTemplateFormatVersion: '2010-09-09'
Resources:
# CloudWatch Alarms for CPU Utilization
CpuUtilizationAlarmHigh:
Type: AWS::CloudWatch::Alarm
Properties:
AlarmName: CpuHighUtilizationAlarm
AlarmDescription: Trigger when CPU Utilization is high
Namespace: AWS/WorkSpaces
MetricName: CPUUtilization
Statistic: Average
Period: 300
EvaluationPeriods: 1
Threshold: 80
ComparisonOperator: GreaterThanOrEqualToThreshold
AlarmActions:
- Ref: TierManagementSNSTopic
CpuUtilizationAlarmLow:
Type: AWS::CloudWatch::Alarm
Properties:
AlarmName: CpuLowUtilizationAlarm
AlarmDescription: Trigger when CPU Utilization is low
Namespace: AWS/WorkSpaces
MetricName: CPUUtilization
Statistic: Average
Period: 300
EvaluationPeriods: 1
Threshold: 20
ComparisonOperator: LessThanOrEqualToThreshold
AlarmActions:
- Ref: TierManagementSNSTopic
# SNS Topic to be triggered by CloudWatch Alarms
TierManagementSNSTopic:
Type: AWS::SNS::Topic
Properties:
DisplayName: WorkSpaceTierManagement
# Lambda function to change WorkSpace tier
WorkSpaceTierChangeLambdaFunction:
Type: AWS::Lambda::Function
Properties:
FunctionName: WorkSpaceTierChange
Handler: index.handler
Role: arn:aws:iam::your-account-id:role/execution_role
Code:
ZipFile: |
import boto3
import json
workspaces_client = boto3.client('workspaces')
def handler(event, context):
sns_message = json.loads(event['Records'][0]['Sns']['Message'])
alarm_name = sns_message.get('AlarmName', 'default-alarm')
# Fetch all WorkSpaces in a specific directory
directory_id = "your-directory-id-here"
workspaces = workspaces_client.describe_workspaces(DirectoryId=directory_id)['Workspaces']
for workspace in workspaces:
workspace_id = workspace['WorkspaceId']
if 'High' in alarm_name:
change_workspace_tier(workspace_id, 'POWERPRO')
elif 'Low' in alarm_name:
change_workspace_tier(workspace_id, 'POWER')
def change_workspace_tier(workspace_id, new_tier):
try:
response = workspaces_client.modify_workspace_properties(
WorkspaceId=workspace_id,
WorkspaceProperties={
'ComputeTypeName': new_tier
}
)
print(f"Changed tier of {workspace_id} to {new_tier}")
except Exception as e:
print(f"Failed to change tier: {e}")
Runtime: python3.8
Timeout: 5
# SNS Subscription to Lambda
SNSTopicSubscription:
Type: AWS::SNS::Subscription
Properties:
Protocol: lambda
TopicArn: !Ref TierManagementSNSTopic
Endpoint: !GetAtt [WorkSpaceTierChangeLambdaFunction, Arn]
The CloudFormation template comprises the following key components:
1. CloudWatch Alarms
Two CloudWatch Alarms are defined:
-
CpuUtilizationAlarmHigh: Triggers when CPU Utilization is above 80%. -
CpuUtilizationAlarmLow: Triggers when CPU Utilization is below 20%.
2. SNS Topic
The TierManagementSNSTopic is an SNS Topic that will receive notifications from the CloudWatch Alarms.
3. Lambda Function
The WorkSpaceTierChangeLambdaFunction is triggered when an SNS notification is received. It changes the tier of WorkSpaces based on the alarm conditions.
4. SNS Subscription
The SNSTopicSubscription subscribes the Lambda function to the SNS Topic, allowing the function to be invoked when an alarm triggers.
Pre-requisites
- AWS CLI installed and configured
- A valid AWS WorkSpaces setup
- IAM role with sufficient permissions for CloudWatch and WorkSpaces (
execution_role)
Deployment Instructions
Follow these steps to deploy the CloudFormation template:
1. Validate the Template
First, validate the CloudFormation template to ensure it is well-formed.
aws cloudformation validate-template --template-body file://path/to/template.yaml
2. Upload the Template to S3 (Optional)
You can upload the template to an S3 bucket if it's too large or if you prefer to keep it centralized.
aws s3 cp path/to/template.yaml s3://your-s3-bucket/
3. Deploy the Stack
You can deploy the CloudFormation stack using the AWS CLI.
# If template is local
aws cloudformation create-stack --stack-name WorkSpacesTierManagement --template-body file://path/to/template.yaml
# Or if template is in S3
aws cloudformation create-stack --stack-name WorkSpacesTierManagement --template-url https://s3.amazonaws.com/your-s3-bucket/template.yaml
4. Monitor the Stack
You can monitor the status of the stack in the AWS CloudFormation console or via the AWS CLI.
aws cloudformation describe-stacks --stack-name WorkSpacesTierManagement
Conclusion
This CloudFormation template offers an automated approach to manage AWS WorkSpaces tiers based on CPU utilization, thus optimizing resource usage. Deploying this stack in your AWS environment can significantly streamline your WorkSpaces management process.
Top comments (0)