Reducing waiting costs in user-level communication

This paper describes a mechanism for reducing the cost of waiting for messages in architectures that allow user-level communication libraries. We reduce waiting costs in two ways: by reducing the cost of servicing interrupts, and by carefully controlling when the system uses interrupts and when it uses polling. We have implemented our mechanism on the SHRIMP multi-computer and integrated it with our user-level sockets library. Experiments show that a hybrid spin-then-block strategy offers good performance in a wide variety of situations, and that speeding up the interrupt path significantly improves performance.