| RATE TYPES is a novel type system to reason about and optimize data-intensive programs. Built around stream languages, RATE TYPES performs static quantitative reasoning about stream rates---the frequency of data items in a stream being consumed, processed, and produced. Despite the fact that streams are fundamentally dynamic, there are two essential concepts of stream rate control---throughput ratio and natural rate---which are intimately related to the program structure itself and can be effectively reasoned about by a type system. RATE TYPES is proven to correspond with a timeaware operational semantics which supports parallelism. The strong correspondence result tolerates arbitrary schedules, and does not require any synchronization between stream filters. RATE TYPES is also implemented on stream programs, demonstrating its effectiveness in predicting stream data rates in real-world stream programs. Applications of RATE TYPES are discussed, including an application of RATE TYPES to optimize energy consumption. |
