Evolution of the Virtual Interface Architecture

T o provide a faster path between applications and the network, most researchers have advocated removing the operating system kernel and its centralized networking stack from the critical path and creating a user-level network interface. With these interfaces, designers can tailor the communication layers each process uses to the demands of that process. Consequently, applications can send and receive network packets without operating system intervention, which greatly decreases communication latency and increases network throughput. Unfortunately, the diversity of approaches and lack of consensus has stalled progress in refining research results into products—a prerequisite to the widespread adoption of these interfaces. Recently, however, Intel, Microsoft, and Compaq have introduced the Virtual Interface Architecture, 1 an emerging standard for cluster or system-area networks. Products based on the VIA have already surfaced, notably GigaNet's GNN1000 network interface (http://www.giganet. com). As more products appear, research into application level issues can proceed and the technology of user-level network interfaces should mature. Several prototypes—among them Cornell University's U-Net 2 —have heavily influenced the VIA. In this article, we describe the architectural issues and design trade-offs at the core of these prototype designs, including • How to provide applications with direct access to the network interface hardware, yet retain sufficient protection to ensure that applications can't interfere with each other. • How to design an efficient, yet versatile programming interface. Applications must be able to access the network interface and still control buffering, scheduling, and addressing. The programming interface, on the other hand, must accommodate a wide variety of hardware implementations. • How to manage resources, in particular memory. Applications must consider the costs of DMA transfers that map a virtual address to a physical one. At the same time, implementation-specific details must be hidden from the application, and the operating system must ultimately control all resource allocation. • How to manage fair access to the network without a kernel path which, in traditional protocol stacks, acts as a central point of control and scheduling. Network performance is traditionally described by the bandwidth achieved when an infinite stream is transmitted. However, an increasing number of applications are more sensitive to the network's round-trip time, or communication latency, and the bandwidth when sending many small messages. Communication latency is due mainly to processing overhead—the time the processor spends handling messages at the sending and receiving ends. This may include managing the buffers, copying the message, computing checksums, handling …