We describe Chromium, a system for manipulating streams of graphics API commands on clusters of workstations. Chromium's stream filters can be arranged to create sort-first and sort-last parallel graphics architectures that, in many cases, support the same applications while using only commodity graphics accelerators. In addition, these stream filters can be extended programmatically, allowing the user to customize the stream transformations performed by nodes in a cluster. Because our stream processing mechanism is completely general, any cluster-parallel rendering algorithm can be either implemented on top of or embedded in Chromium. In this paper, we give examples of real-world applications that use Chromium to achieve good scalability on clusters of workstations, and describe other potential uses of this stream processing technology. By completely abstracting the underlying graphics architecture, network topology, and API command processing semantics, we allow a variety of applications to run in different environments.

It has become increasingly difficult to drive a modern highperformance graphics accelerator at full speed with a serial immediate -mode graphics interface. To resolve this problem, retainedmode 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 highperformance graphics applications.

OpenGL's window system support for the X Window System explicitly allows implementations to support direct rendering of OpenGL commands to the graphics hardware. Rendering directly to the hardware avoids the overhead of packing and relaying protocol requests to the X server inherent in indirect rendering. The OpenGL implementation available for Silicon Graphics workstations supports direct rendering using virtualizable graphics hardware in conjunction with the kernel and the X server. The techniques described provide "maximum performance" rendering for OpenGL. Some of the issues are specific to OpenGL, but most of the techniques described are appropriate for the implementation of any high-performance direct rendering graphics interface. Keywords: OpenGL, Virtual Graphics,Direct Rendering. 1 Introduction The OpenGL graphics system [14, 11] is a window system independent software interface to graphics hardware for 3D rendering. GLX [8] is the OpenGL extension to the X Window System that...

Abstract not found

This thesis develops an engineering practice and design methodology to enable us to use CMOS analog VLSI chips to perform more accurate and precise computation. These techniques form the basis of an approach that permits us to build computer graphics and neural network applications using analog VLSI. The nature of the design methodology focuses on defining goals for circuit behavior to be met as part of the design process. To increase the accuracy of analog computation, we develop techniques for creating compensated circuit building blocks, where compensation implies the cancellation of device variations, offsets, and nonlinearities. These compensated building blocks can be used as components in larger and more complex circuits, which can then also be compensated. To this end, we develop techniques for automatically determining appropriate parameters for circuits, using constrained optimization. We also fabricate circuits that implement multi-dimensional gradient estimation for a grad...

To support OpenGL TM and PEX rendering within the Silicon Graphics X server without compromising interactivity, we devised and implemented a scheme named multi-rendering. Making minimal changes to the X Consortium sample server's overall structure, the scheme allows independent processes within the X server's address space to perform OpenGL rendering asynchronously to the X server's main thread of execution. The IRIX operating system's process share group facility, user-level and pollable semaphores, and support for virtualized direct access rendering are all leveraged to support multi-rendering. The Silicon Graphics implementation of PEX also uses the multi-rendering facility and works by converting rendering requests into OpenGL commands. Multi-rendering is contrasted with other schemes for improving server interactivity. Unlike co-routines, multi-rendering supports multi-processing; unlike multi-threading, multi-rendering requires minimal locking overhead. Introduction The curre...

This paper summarises the author's views on current developments in interactive scientific visualisation. It is based on a talk presented at the Eurographics '89 conference, held in Hamburg in September 1989. The paper takes issue with the direction of some current work and identifies areas where new ideas are needed. It has three main sections: data presentation methods, current visualisation system architectures, and a new approach based on parallel processing. 1 Introduction The upsurge of interest in visualisation was given a major impetus by a report prepared for the National Science Foundation in the USA (the ViSC report) [16]. The main thrust of this was to examine how the USA could remain competitive in this area, and therefore what research should be funded by the government. A major problem identified was how researchers could assimilate the truly vast amounts of data being poured out by supercomputers --- what the report termed "firehoses of data". The report recommended a s...

Communication forms the backbone of parallel graphics, allowing multiple functional units to cooperate to render images. The cost of this communication, both in system resources and money, is the primary limit to parallelism. We examine the use of object and image parallelism and describe architectures in terms of the sorting communication that connects these forms of parallelism. We introduce an extended taxonomy of parallel graphics architecture that more fully distinguishes architectures based on their sorting communication, paying particular attention to the difference between sorting fragments after rasterization, and sorting samples after fragments are merged with the framebuffer. We introduce three new forms of communication, distribution, routing and texturing, in addition to sorting. Distribution connects object parallel pipeline stages, routing connects image parallel pipeline stages, and texturing connects untextured fragments with texture memory. All of these types of communication allow the parallelism of successive pipeline stages to be decoupled, and thus load-balanced. We generalize communication to include not only interconnect, which provides communication across space, but also memory, which functions as communication across time. We examine a number of architectures from this communicationcentric perspective, and discuss the limits to their scalability. We draw conclusions to the limits of both image parallelism and broadcast communication and suggest architectures that avoid these limitations.

The modern graphics processing unit (GPU) is the result of 40 years of evolution of hardware to accelerate graphics processing operations. It represents the convergence of support for multiple market segments: computer-aided design, medical imaging, digital content creation, document and presentation applications, and entertainment applications. The exceptional performance characteristics of the GPU make it an attractive target for other application domains. We examine some of this evolution, look at the structure of a modern GPU, and discuss how graphics processing exploits this structure and how nongraphical applications can take advantage of this capability. We discuss some of the technical and market issues around broader adoption of this technology.