Budgets and cost anomaly detection in AWS - Mini Project: Build & Apply

Budgets and Cost Anomaly Detection Setup in AWS

📖 Scenario: You are managing cloud costs for a small company. To keep spending under control, you want to set a monthly budget and get alerts if costs go over the budget or if unusual spending happens.

🎯 Goal: Build an AWS Budgets configuration with a monthly budget and enable cost anomaly detection alerts to monitor unexpected cost spikes.

📋 What You'll Learn

Create an AWS Budget with a fixed monthly amount of $500

Set the budget to track actual costs

Add an alert to notify when costs exceed 80% of the budget

Enable AWS Cost Anomaly Detection for the linked account

Configure anomaly detection to send alerts to an SNS topic

💡 Why This Matters

🌍 Real World

Companies use budgets and anomaly detection to control cloud spending and avoid surprises in their bills.

💼 Career

Cloud engineers and cost managers need to set budgets and monitor anomalies to optimize cloud costs and alert stakeholders.

1

Create the AWS Budget dictionary

Create a dictionary called budget with these exact keys and values: BudgetName set to "MonthlyCostBudget", BudgetLimit set to 500, and TimeUnit set to "MONTHLY".

AWS

# Your code here to create the budget dictionary

Hint

Use a Python dictionary with keys exactly as BudgetName, BudgetLimit, and TimeUnit.

2

Add alert threshold to the budget

Add a key Alerts to the budget dictionary. Set it to a list with one dictionary that has Threshold set to 80, ComparisonOperator set to "GREATER_THAN", and NotificationType set to "ACTUAL".

AWS

budget = {
    "BudgetName": "MonthlyCostBudget",
    "BudgetLimit": 500,
    "TimeUnit": "MONTHLY"
}
# Add Alerts key with threshold dictionary here

Hint

Alerts is a list with one dictionary inside. Make sure to use the exact keys and values.

3

Create the anomaly detection configuration

Create a dictionary called anomaly_detection_config with keys: AccountId set to "123456789012", MonitorName set to "MonthlyCostAnomalyMonitor", and SnsTopicArn set to "arn:aws:sns:us-east-1:123456789012:CostAnomalyAlerts".

AWS

budget = {
    "BudgetName": "MonthlyCostBudget",
    "BudgetLimit": 500,
    "TimeUnit": "MONTHLY",
    "Alerts": [
        {
            "Threshold": 80,
            "ComparisonOperator": "GREATER_THAN",
            "NotificationType": "ACTUAL"
        }
    ]
}

# Create anomaly_detection_config dictionary here

Hint

Use the exact keys and values for the anomaly detection configuration dictionary.

4

Complete the AWS Budgets and Anomaly Detection setup

Add a key CostFilters to the budget dictionary and set it to a dictionary with LinkedAccount key set to a list containing "123456789012". Then add a key EnableAnomalyDetection set to True in anomaly_detection_config.

AWS

budget = {
    "BudgetName": "MonthlyCostBudget",
    "BudgetLimit": 500,
    "TimeUnit": "MONTHLY",
    "Alerts": [
        {
            "Threshold": 80,
            "ComparisonOperator": "GREATER_THAN",
            "NotificationType": "ACTUAL"
        }
    ]
}

anomaly_detection_config = {
    "AccountId": "123456789012",
    "MonitorName": "MonthlyCostAnomalyMonitor",
    "SnsTopicArn": "arn:aws:sns:us-east-1:123456789012:CostAnomalyAlerts"
}

# Add CostFilters to budget and EnableAnomalyDetection to anomaly_detection_config here

Hint

CostFilters should filter by the linked account ID. Enable anomaly detection by setting the key to True.

Practice

1. What is the main purpose of AWS Budgets in cloud cost management?

easy

A. To automatically reduce cloud service prices

B. To create backups of billing data

C. To set spending limits and receive alerts when costs approach those limits

D. To monitor network traffic between AWS services

Solution

1. Step 1: Understand AWS Budgets function

   AWS Budgets allow users to set cost or usage limits and get alerts when those limits are near or exceeded.

2. Step 2: Compare options to this function

   Only To set spending limits and receive alerts when costs approach those limits describes setting limits and alerts, which matches AWS Budgets' purpose.

3. Final Answer:

   To set spending limits and receive alerts when costs approach those limits -> Option C

4. Quick Check:

   AWS Budgets = Spending limits and alerts [OK]

Hint: Budgets set limits and alert you before overspending [OK]

Common Mistakes:

• Thinking Budgets automatically reduce costs
• Confusing Budgets with backup or network tools
• Assuming Budgets monitor traffic instead of costs

2. Which of the following is the correct way to create a cost anomaly detection monitor using AWS CLI?

easy

A. aws ec2 create-anomaly-monitor --name MyMonitor

B. aws ce create-budget --name MyMonitor

C. aws ce start-anomaly-monitor --monitor-name MyMonitor

D. aws ce create-anomaly-monitor --monitor-name MyMonitor

Solution

1. Step 1: Identify correct AWS CLI command for anomaly monitor

   The AWS Cost Explorer (ce) service uses 'create-anomaly-monitor' to create anomaly monitors.

2. Step 2: Check options for correct syntax and service

   aws ce create-anomaly-monitor --monitor-name MyMonitor uses 'aws ce create-anomaly-monitor' with correct parameter '--monitor-name'. Others use wrong commands or services.

3. Final Answer:

   aws ce create-anomaly-monitor --monitor-name MyMonitor -> Option D

4. Quick Check:

   Correct CLI command for anomaly monitor creation = aws ce create-anomaly-monitor --monitor-name MyMonitor [OK]

Hint: Use 'aws ce create-anomaly-monitor' to create monitors [OK]

Common Mistakes:

• Using 'create-budget' instead of anomaly monitor command
• Using EC2 service commands for cost anomaly detection
• Incorrect parameter names like '--name' instead of '--monitor-name'

3. Given this AWS CLI command output snippet for a cost anomaly detection alert:

{
  "AnomalyId": "12345",
  "AnomalyScore": 85,
  "Service": "Amazon S3",
  "StartTime": "2024-05-01T00:00:00Z",
  "EndTime": "2024-05-01T23:59:59Z"
}

What does the AnomalyScore of 85 indicate?

medium

A. A high likelihood of cost anomaly

B. A moderate likelihood of cost anomaly

C. A low likelihood of cost anomaly

D. An error in anomaly detection

Solution

1. Step 1: Understand AnomalyScore meaning

   AnomalyScore ranges from 0 to 100; higher scores mean higher likelihood of unusual cost behavior.

2. Step 2: Interpret score 85

   Score 85 is close to 100, indicating a high likelihood of a cost anomaly.

3. Final Answer:

   A high likelihood of cost anomaly -> Option A

4. Quick Check:

   High AnomalyScore means high anomaly chance [OK]

Hint: Higher AnomalyScore means stronger anomaly signal [OK]

Common Mistakes:

• Thinking 85 is low or moderate score
• Confusing AnomalyScore with error codes
• Ignoring the score scale from 0 to 100

4. You created a budget with this AWS CLI command:

aws budgets create-budget --account-id 123456789012 --budget '{"BudgetName":"MyBudget","BudgetLimit":{"Amount":"100","Unit":"USD"},"TimeUnit":"MONTHLY","BudgetType":"COST"}'

But you receive an error. What is the most likely cause?

medium

A. Missing required parameter --notifications-with-subscribers

B. Incorrect JSON format for the budget parameter

C. Using --account-id instead of --account-id is invalid

D. BudgetLimit Amount must be an integer, not a string

Solution

1. Step 1: Check JSON format in --budget parameter

   The JSON string uses double quotes inside single quotes but may have formatting issues or missing escapes.

2. Step 2: Identify common CLI JSON errors

   Incorrect JSON formatting or missing escapes cause CLI errors; the command likely fails due to this.

3. Final Answer:

   Incorrect JSON format for the budget parameter -> Option B

4. Quick Check:

   JSON format errors cause CLI command failures [OK]

Hint: Validate JSON syntax carefully in CLI commands [OK]

Common Mistakes:

• Assuming missing notifications cause errors
• Confusing parameter names like --account-id
• Thinking Amount must be integer, but string is allowed

5. You want to monitor your AWS costs and get notified only when unusual spending spikes happen in specific services tagged with Environment=Production. Which combination of AWS features should you use?

hard

A. Create a cost anomaly detection monitor filtered by tag and set SNS notifications for alerts

B. Create a budget with a fixed limit and enable email alerts for all costs

C. Use AWS CloudTrail to track cost changes and send alerts via Lambda

D. Set up AWS Config rules to monitor cost and trigger alarms

Solution

1. Step 1: Identify feature for unusual spending spikes

   Cost anomaly detection monitors unusual cost spikes, suitable for this use case.

2. Step 2: Use tags to filter specific services and set notifications

   Filtering by tag 'Environment=Production' and using SNS notifications matches the requirement.

3. Final Answer:

   Create a cost anomaly detection monitor filtered by tag and set SNS notifications for alerts -> Option A

4. Quick Check:

   Cost anomaly + tag filter + SNS alerts = Create a cost anomaly detection monitor filtered by tag and set SNS notifications for alerts [OK]

Hint: Use anomaly detection with tag filters and SNS for targeted alerts [OK]

Common Mistakes:

• Using budgets which alert on fixed limits, not anomalies
• Confusing CloudTrail or Config with cost anomaly detection
• Ignoring tag filtering for specific environment costs