The Pismo platform uses rate limiting to maintain stable performance and prevent service degradation during peak usage. Rate limiting sets the maximum number of simultaneous requests allowed through an API. If requests exceed the limit, the platform returns error HTTP 429 (Too Many Requests), as defined in RFC 6585, Section 4.
Pismo adds rate limits to APIs once we identify performance needs.
Preparing for high volume
If you expect to process a high volume of simultaneous requests, work with Pismo to define the best approach. Create a Pismo service desk ticket to get the conversation started.
Best practices
To keep from experiencing the rate limit error, follow these best practices:
- Prohibit load testing in the production environment is prohibited.
- Create a retry mechanism with an exponential backoff for when an error code is encountered. This involves creating a series of retry statements that send a decreasing amount of requests, spaced out in exponentially growing increments. For example, you can send a retry request after 2 minutes, then after 4 minutes, then after 8 minutes, and so on. Once you have determined an appropriate retry interval, follow this pattern until all requests are handled successfully.