Amazon monitoring bandwidth usage
This knowledge base provides details about the bandwidth usage of the Amazon monitor and its child monitors. Both ingress (download) and egress (upload) bandwidth are listed for each operation.
Note: The bandwidth values mentioned below are sample estimates. Actual bandwidth consumption may vary depending on the customer’s AWS resources, configuration, and the amount of data collected. Credential validation
- This API is used to validate the provided AWS credentials.
|
API Host
|
Ingress (KB)
|
Egress (KB)
|
|
sts-global-default.us-east-1.amazonaws.com
|
5.2
|
1.5
|
Region validation (Optional)
- This API is used to validate the selected AWS regions and optimize subsequent API calls.
| Host | Ingress (KB) | Egress (KB) |
| ec2.us-east-1.amazonaws.com | 6.8 | 1.5 |
Amazon monitor discovery
Scenario 1: Discovery with 1 RDS, 1 EKS & 1 Lambda in a single region
- The following bandwidth usage applies when discovering one RDS instance, one EKS cluster, and one Lambda function in a single AWS region for an Amazon monitor.
- Both the credential validation & region validation is done once per discovery.
| Host | Ingress (KB) | Egress (KB) |
| sts-global-default.us-east-1.amazonaws.com | 5.2 | 1.5 |
| ec2.us-east-1.amazonaws.com | 6.8 | 1.5 |
| rds.us-west-1.amazonaws.com | 6.6 | 1.6 |
| lambda.us-west-1.amazonaws.com | 6 | 1 |
| eks.us-west-1.amazonaws.com | 5.3 | 1 |
| Total Bandwidth | 29.9 | 6.6 |
Scenario 2: Discovery with 1 RDS in a single region
- The following bandwidth usage applies when discovering one RDS instance in a single AWS region for an Amazon monitor.
| Host | Ingress (KB) | Egress (KB) |
| sts-global-default.us-east-1.amazonaws.com | 5.2 | 1.5 |
| ec2.us-east-1.amazonaws.com | 6.8 | 1.5 |
| rds.us-west-1.amazonaws.com | 6.6 | 1.6 |
| Total Bandwidth | 18.6 | 4.6 |
Data Collection
RDS Instance data collection
- The following bandwidth usage applies to data collection for an RDS Aurora MySQL instance monitor in a single poll.
| Host | Ingress (KB) | Egress (KB) |
| sts-global-default.us-east-1.amazonaws.com | 5.2 | 1.5 |
| rds.us-west-1.amazonaws.com | 12.8 | 3.2 |
| monitoring.us-west-1.amazonaws.com | 12.1 | 19.5 |
| Total Bandwidth | 30.1 | 24.2 |
Lambda data collection
- The following bandwidth usage applies to data collection for a Lambda monitor in a single poll.
| Host | Ingress (KB) | Egress (KB) |
| sts-global-default.us-east-1.amazonaws.com | 5.2 | 1.5 |
| lambda.us-west-1.amazonaws.com | 7.5 | 1.1 |
| monitoring.us-west-1.amazonaws.com | 5.4 | 8.3 |
| Total Bandwidth | 18.1 | 10.9 |
Elastic Kubernetes Service (EKS) data collection
- The following bandwidth usage applies to an EKS cluster monitor during a single poll with the below resources:
- 1 node
- 4 pods
- 5 containers
- 3 services
- 1 deployment
| Host | Ingress (KB) | Egress (KB) |
| AWS APIs | 71.5 | 40.7 |
| lambda.us-west-1.amazonaws.com | 9.4 | 1.2 |
| monitoring.us-west-1.amazonaws.com | 162.5 | 26 |
| Total Bandwidth | 243.4 | 67.9 |
Calculation summary
Example
- Amazon discovery
- Single request to fetch all the RDS instances in the account.
| Ingress (KB): 50 * 6.6 (RDS API) + 5.2 (Credential validation) + 6.8 (Region validation) = 342 KB Egress (KB): 1.6 (RDS API) + 1.5 (Credential validation) + 1.5 (Region validation) = 4.6 KB |
| Ingress (KB): 342 * 288 polls = 98,496 KB (~96.2 MB) Egress (KB): 4.6 × 288 polls = 1,325 KB (~1.29 MB) Poll Interval: 5 mins - 288 polls per day |
- Data collection
- Each child monitors are polled separately. Region validation is not done during data collection.
| Ingress (KB): 50 * 30.1 (including credential validation) = 1505 KB Egress (KB): 50 * 24.2 (including credential validation) = 1210 KB |
| Ingress (KB): 1,505 × 288 = 433,440 KB (~423.3 MB) Egress (KB): 1,210 × 288 = 348,480 KB (~340.3 MB) Poll Interval: 5 mins - 288 polls per day |
Summary
- Amazon monitor discovery
- Discovery bandwidth depends on the number of services, regions, and resources per region.
- In this example, bandwidth is calculated for RDS, EKS, and Lambda in a single region, excluding S3, SNS, and Billing.
- Child monitor data collection
- Each discovered resource is added as a child monitor and collects configuration and CloudWatch metrics at the configured poll interval.
- Bandwidth varies by service based on the resource type, metrics collected, and associated child entities.
Additional References
New to ADSelfService Plus?
Related Articles
Cloud Monitoring API Usage
How Cloud Monitoring Works? Cloud Monitoring involves collecting, analyzing, and visualizing performance metrics and telemetry data from cloud infrastructure and applications via APIs. Cloud Monitoring data collection flow: API Authentication: ...Memory Usage Calculation (Linux- Telnet/SSH)
Physical Memory Utilization Calculation: Note: For version 16800 and above , the Skip option will be under Settings -> Performance Polling -> Servers -> Linux monitor type. Formula: (i) If Skip buffer and cache from monitoring (Applicable only for ...Monitoring HAProxy - Configuring the HAProxy Stats page
To effectively monitor an HAProxy instance, you can enable and access the HAProxy stats page, which provides real-time metrics about HAProxy's performance. Steps to Enable HAProxy Stats: Open the HAProxy Configuration File: The HAProxy configuration ...Add Amazon Monitor with Required IAM Permissions
This article explains how to add Amazon monitors using only the required APIs (Custom IAM Policy - second method ↓) Methods to Configure IAM Permissions for Amazon Monitoring Read Only Access (AWS Managed Policy - Recommended) This predefined policy ...Optimizing AWS API Usage to Avoid Rate Limit Errors in Applications Manager
If you encounter the error Request Limit Exceeded This is not an issue caused by Applications Manager but rather a result of AWS API rate limits. These limits are set by AWS to ensure fair usage across all customers. The error occurs when the ...