Extensive caching is a key feature of the Echo distributed file system. Echo client machines maintain coherent caches of file and directory data and properties, with write-behind (delayed write-back) ofallcached information. Echo specifies ordering constraints on this write-behind, enabling applications to store and maintain consistent data structures in the file system even when crashes or network faults prevent some writes from being completed. In this paper we describe the Echo cache's coherence and ordering semantics, show how they can improve the performance and consistency of applications, explain how they are implemented. We also discuss the general problem of reliably notifying applications and users when write-behind is lost; we addressed this problem as part of the Echo design, but did not find a fully satisfactory solution.

Theoretical and experimental results have indicated that automatic decompositions can discover modest amounts of parallelism. These investigations have tended to ignore the practical problems of language run-time organization, such as synchronization, communication, memory organization, resource management, and input/output. This paper describes a language implementation effort which combines the investigation of implicit and explicit parallel decomposition facilities with the practical considerations of system organization on a multiprocessor computer, Cm*.