The design of a parallel graphics interface

Abstract

It has become increasingly difficult to drive a modern high-performance graphics accelerator at full speed with a serial immediate-mode graphics interface. To resolve this problem, retained-mode constructs have been integrated into graphics interfaces. While retained-mode constructs provide a good solution in many cases, at times they provide an undesirable interface model for the application programmer, and in some cases they do not solve the performance problem. In order to resolve some of these cases, we present a parallel graphics interface that may be used in conjunction with the existing API as a new paradigm for high-performance graphics applications. The parallel API extends existing ideas found in OpenGL and X11 that allow multiple graphics contexts to simultaneously draw into the same image. Through the introduction of synchronization primitives, the parallel API allows parallel traversal of an explicitly ordered scene. We give code examples which demonstrate how the API can be used to expose parallelism while retaining many of the desirable features of serial immediate-mode programming. The viability of the API is demonstrated by the performance of our implementation which achieves scalable performance on a 24 processor system.