Aggregate and Visualize Comprehensive RDS CPU Utilization

import boto3 from datetime import datetime, timedelta region_name=None # None when you want to run the script for all regions #last_n_days = 30 # AWS Credentials - replace with your method to retrieve AWS credentials creds = _get_creds(cred_label)['creds'] # Placeholder function access_key = creds['username'] secret_key = creds['password'] def get_aws_regions(): """Get a list of all AWS regions.""" ec2 = boto3.client('ec2',aws_access_key_id=access_key,aws_secret_access_key=secret_key,region_name='us-east-1') regions = ec2.describe_regions() return [region['RegionName'] for region in regions['Regions']] def fetch_rds_instances(region): """Fetch all RDS instances in a specific region.""" rds = boto3.client('rds',aws_access_key_id=access_key,aws_secret_access_key=secret_key, region_name=region) paginator = rds.get_paginator('describe_db_instances') page_iterator = paginator.paginate() rds_instances = [] for page in page_iterator: for instance in page['DBInstances']: rds_instances.append({ 'DBInstanceIdentifier': instance['DBInstanceIdentifier'], 'Region': region, }) return rds_instances def fetch_cpu_utilization(db_instance_identifier, region, start_time, end_time): """Fetch the average CPU utilization for an RDS instance.""" cloudwatch = boto3.client('cloudwatch',aws_access_key_id=access_key,aws_secret_access_key=secret_key, region_name=region) metrics = cloudwatch.get_metric_statistics( Namespace='AWS/RDS', MetricName='CPUUtilization', Dimensions=[{'Name': 'DBInstanceIdentifier', 'Value': db_instance_identifier}], StartTime=start_time, EndTime=end_time, Period=3600, Statistics=['Average'] ) data_points = metrics.get('Datapoints', []) if data_points: avg_cpu = sum(dp['Average'] for dp in data_points) / len(data_points) else: avg_cpu = 0 return avg_cpu def plot_cpu_utilization(region_name=None, last_n_days=7): """Plot CPU utilization for RDS instances.""" start_time = datetime.utcnow() - timedelta(days=last_n_days) end_time = datetime.utcnow() regions = [region_name] if region_name else get_aws_regions() for region in regions: rds_instances = fetch_rds_instances(region) avg_utilizations = [] for instance in rds_instances: avg_cpu = fetch_cpu_utilization(instance['DBInstanceIdentifier'], region, start_time, end_time) avg_utilizations.append((instance['DBInstanceIdentifier'], avg_cpu)) avg_utilizations.sort(key=lambda x: x[1], reverse=True) top_instances = avg_utilizations[:3] bottom_instances = avg_utilizations[-3:] instance_ids = [x[0] for x in top_instances + bottom_instances] utilizations = [x[1] for x in top_instances + bottom_instances] # Plotting context.plot.add_trace( name="CPU Utilization", xpts=instance_ids, ypts=utilizations, tracetype='bar' ) context.plot.xlabel = 'Instance ID' context.plot.ylabel = 'Average CPU Utilization (%)' context.plot.title = f'Top & Bottom 3 RDS Instances by CPU Utilization (Last {last_n_days} Days)' # Example usage plot_cpu_utilization(region_name, last_n_days) # For all regions # plot_cpu_utilization(region_name='us-east-1', last_n_days=30) # For a specific region