A method that includes establishing an open connection for responding to requests from clients supported by an application server. The method may further include establishing a set of queues configured for storing requests received from the client via the open connection. The method may further include selecting requests from the queues based on a rate limit threshold and burst limit threshold of the application server. The rate limit threshold may refer to a number of requests that the application server can process within a first time duration, while the burst limit threshold may refer to a number of requests that the application server can process within a second time duration that is shorter than the first time duration. The method may further include transmitting the requests to a set of data processing servers connected to the application server and receiving an indication that the requests have been processed.