A Distributed Shared Memory (DSM) system is a software or hardware mechanism that provides a distributed application with a shared virtual address space. The efficiency of a DSM system relies mainly on a memory coherency protocol and an efficient communication layer. This article proposes a design for implementing the communication layer on top of the Virtual Interface Architecture (VIA), an industry standard for user-level networking protocols on high-speed clusters. User-level communication protocols operate in a user mode, thus removing the operating system kernel?s overhead from the critical communication pass and significantly diminishing communication overhead as a result. We analyze VIA?s facilities and limitations in order to ascertain which implementation trade-offs can be best applied to our development of an efficient communication substrate optimized for DSM requirements. We then implement a multithreaded version of the home-based lazy release consistency (HLRC) protocol on top of this efficient substrate. We evaluate and analyze the performance of this protocol over a wide set of benchmark applications.