NPU-Based Image Compositing in a Distributed Visualization System

Abstract

This paper describes the first use of a Network Processing Unit (NPU) to perform hardware-based image composition in a distributed rendering system. The image composition step is a notorious bottleneck in a clustered rendering system. Furthermore, image compositing algorithms do not necessarily scale as data size and number of nodes increase. Previous researchers have addressed the composition problem via software and/or custom-built hardware. We used the heterogeneous multicore computation architecture of the Intel IXP28XX NPU, a fully programmable commercial off-the-shelf (COTS) technology, to perform the image composition step. With this design, we have attained a nearly four-times performance increase over traditional software-based compositing methods, achieving sustained compositing rates of 22-28 fps on a 1,024 x 1,024 image. This system is fully scalable with a negligible penalty in frame rate, is entirely COTS, and is flexible with regard to operating system, rendering software, graphics cards, and node architecture. The NPU-based compositor has the additional advantage of being a modular compositing component that is eminently suitable for integration into existing distributed software visualization packages.

Massively parallel volume rendering using 2--3 swap image compositing

“…More recently, Nonaka et al [12] presented a hybrid image composition method for sort-last parallel rendering on graphics clusters with the MPC image compositor, which offers high-performance image compositing at a reasonable cost. Pugmire et al [13] presented a new solution that utilizes a network processing unit (NPU) for hardware-based image compositing in a distributed rendering system. Nevertheless, software image compositing is still widely used in most applications to date.…”

An Efficient Co-processing Framework for Large-Scale Scientific Applications

“…Subsequently, the overall process of scientific discovery is optimized by reducing both data transfer and storage cost early in the data analysis pipeline. Some related work [7] [6] have demonstrated such runtime visualization of simulation on a parallel computer, but the scales of the system and the problem were fairly small. Simulation-time visualization also gives the scientists an option to visually monitor the simulation while it is running.…”