Serverless lessons learned #7 rate limiting
Download as PPTX, PDF0 likes16 views
The document discusses rate limiting in AWS, emphasizing its necessity for managing incoming requests to prevent errors like 'requestlimitexceeded.' It highlights that while rate limits cannot be avoided, services should be adapted to handle them effectively, with AWS SDK providing built-in management options. The author encourages defining and testing custom rate limits for APIs as part of service requirements.
1 of 12
Download to read offline
Serverless lessons learned #7 rate limiting
- 1. Rate Limiting ? Serverless on AWS lessons learned #7 Are you really production-ready?
- 2. Rate Limiting (or Throttling) is a widely adopted pattern to limit the amount of incoming requests.
- 3. Nearly every AWS API call is rate limited in some way. Sooner or later you will see something like this: RequestLimitExceeded
- 4. You will encounter rate limiting using the management API* and using a service API** * e.g. create DynamoDB tables **e.g. read data from a table
- 5. Some rate limits are fixed and some can change over time*. *e.g. throughput limits until autoscaling reacts.
- 7. You can’t avoid rate limiting. You should minimize the occurrences and adapt your services to handle them properly!
- 8. AWS SDK already handles rate limiting and retries after a short delay*. You can customize this behavior if needed. *see description for details
- 9. For 3rd party APIs with rate limiting, think about implementing a retry strategy with exponential backoff.
- 10. Define, implement and test rate limits for your own APIs*. Make them part of your service requirements. *see description for custom rate limits with AWS API-Gateway
- 11. Rate Limiting ?
- 12. thumbs up, comment, share What are your thoughts? Is anything important missing? If you liked it, please leave a comment. connect with me!