Matej Šnuderl (2018) Rate limiting in API management. EngD thesis.
Abstract
With ever growing usage of World Wide Web, number of requests to web APIs is increasing rapidly. DoS attacks and service abuses are becoming easier to execute, and more common every day. Quality of service is becoming more important as competition is rising. To build robust and reliable services, software engineers have to take this into account when designing web APIs, to deliver end users with a pleasant and reliable experience. In this thesis we delve into rate limiting in web API management to deal with those problems on scale. We propose an approach to rate limiting when request weighting is key, and cannot be estimated/calculated upfront. We show how integration of such approach into a real working system can help in achieving high stability and performance improvements, while unlocking some advanced API monetisation opportunities.
Actions (login required)