Pupa: A Low-Latency Communication System for Fast Ethernet

Pupa is a low-latency communication system that provides the same quality of message delivery as TCP but is designed speci cally for a parallel computing cluster connected by a 100 Mbits/sec Fast Ethernet. The implementation has been operational for over a year, and several systems have been built on top of Pupa, including a compiler-directed distributed shared virtual memory system, Locust , and a parallel multimedia index server, PAMIS . To minimize bu er management overhead, Pupa uses per-sender/per-receiver xedsized FIFO bu ers to optimize for the common case, rather than a shared variable-length linked-list bu er pool. In addition, Pupa features a sender-controlled acknowledgement and an optimistic ow control scheme to reduce the overhead of providing reliable in-order message delivery. Our performance results show that Pupa is twice as fast as the fast path of TCP in terms of latency, and is about 1.5 times better in terms of throughput. This paper presents the design decisions during the development of Pupa, the results of a detailed performance study of the Pupa prototype, as well as the implementation experiences from Pupabased applications development.