In computer graphics, rendering is the process by which an abstract description of a scene is converted to an image. When the scene is complex, or when high-quality images or high frame rates are required, the rendering process becomes computationally demanding. To provide the necessary levels of performance, parallel computing techniques must be brought to bear. Although parallelism has been exploited in computer graphics since the early days of the field, its initial use was primarily in specialized applications. The VLSI revolution of the late 1970Õs and the advent of scalable parallel computers during the late 1980Õs changed this situation. Today, parallel hardware is routinely used in graphics workstations, and numerous software-based rendering systems have been developed for general-purpose parallel architectures. This article provides a broad introduction to the subject of parallel rendering, encompassing both hardware and software systems. The focus is on the underlying concepts and the issues which arise in the design of parallel rendering algorithms and systems. We examine the different types of parallelism and how they can be applied in rendering applications. Concepts from parallel computing, such as data decomposition, task granularity, scalability, and load balancing, are considered in relation to the rendering

Distribution ray tracing uses Monte Carlo integration to solve the rendering equation. This technique was introduced by Cook et. al, and was notable because of its simplicity and its ability to simulate areal luminaires, camera lens e ects, motion blur, and imperfect specular re ection[5]. Distribution

Global illumination is an area of research which tries to develop algorithms and methods to render images of

Parallelising ray tracing with a data parallel approach allows rendering of arbitrarily large models, but the inherent load imbalances may lead to severe inefficiencies. To compensate for the uneven load distribution, demand-driven tasks may be split off and scheduled to processors that are less busy. We propose a hybrid scheduling algorithm which brings tasks and data together according to coherence between rays. Coherent tasks are scheduled demand driven and the remainder is executed data parallel. This method removes the worst hot-spots from the data parallel component and reschedules those as demand driven tasks, thereby evening out the workload. Processing power, communication and memory are three resources which should be evenly used. Our current implementation is assessed against these requirements. Related issues, such as the distribution of the workload over space and the resulting requirements for the distribution objects over the processors, are investigated as well. Final...

We describe a new data parallel algorithm for raytracing. Load balancing is achieved through the use of processor allocation, which continually remaps available resources. In this manner heterogeneous data bases are handled without the usual problems of low resource usage. The proposed approach adapts well to both extremes: a small number of rays and a large database; a large number of rays and a small database. The algorithm scales linearly---over a wide range---in the number of rays and available processors. We present an implementation on the Connection Machine CM2 system and provide timings. R'esum'e Cet article pr'esente un nouvel algorithme parall`ele pour le lancer de rayons. L'allocation des processeurs, qui distribue la tache aux resources disponibles, permet de garder une charge bien r'epartie. Ainsi 'evitons nous les probl`emes usuels dus aux ressources de bas niveau tout en manipulant des structures de donn'ees h'et'erog`enes. Notre approche s'applique aussi bien `a un f...

Volume rendering is a useful but cpu-intensive method for visualizing large scalar elds. The time to render a single image may be reduced by parallel processing. This paper reports on performance experiments with the StormView volume renderer, which is parallelized on a set of 57 MIPS / 17 MFLOPS workstations connected by a 10 Mbps Ethernet. For certain user patterns, we show that our parallelization exhibits substantial speedups. We compare the performance of a dynamic and a static load balancing algorithm. 1

We present a refined approach to parallel array fusion that uses indexed types to specify the internal representation of each array. Our approach aids the client programmer in reasoning about the performance of their program in terms of the source code. It also makes the intermediate code easier to transform at compile-time, resulting in faster compilation and more reliable runtimes. We demonstrate how our new approach improves both the clarity and performance of several end-user written programs, including a fluid flow solver and an interpolator for volumetric data. Categories and Subject Descriptors D.3.3 [Programming Languages]: Language Constructs and Features—Concurrent programming structures; Polymorphism; Abstract data types

This thesis presents and analyzes scalable algorithms for dynamic load balancing and mapping in distributed computer systems. The algorithms are distributed and concurrent, have no central thread of control, and require no centralized communication. They are derived using spectral properties of graphs: graphs of physical network links among computers in the load balancing problem, and graphs of logical communication channels among processes in the mapping problem. A distinguishing characteristic of these algorithms is that they are scalable: the expected cost of execution does not increase with problem scale. This is proven in a scalability theorem which shows that, for several simple disturbance models, the rate of convergence to a solution is independent of scale. This property is extended through simulated examples and informal argument to general and random disturbances. A worst case disturbance is presented and shown to occur with vanishing probability as the problem scale increas...