high performance computing on graphics processing units: hgpu.org

Rendering massive 3D models has been recognized as a challenging task. Due to the limited size of GPU memory, a massive model containing hundreds of millions of primitives cannot fit into most of modern GPUs. By applying parallel levelof-detail (LOD), as proposed in [1], only a portion of primitives instead of the whole are necessary to be streamed to the GPU. However, the low bandwidth in CPU-GPU communication is still the major bottleneck that prevents users from achieving highperformance rendering of massive 3D models on a single-GPU system. This paper explores a device-level parallel design that distributes the workloads for both GPU out-of-core and LOD processing in a multi-GPU multi-display system. Our multi-GPU out-of-core takes advantages of a load-balancing method and seamlessly integrates with the parallel LOD algorithm. By using frame-to-frame coherence, the overhead of data transferring is significantly reduced on each GPU. Our experiments show a highly interactive visualization of the "Boeing 777" airplane model that consists of over 332 million triangles and over 223 million vertices.