Search

Parallel rendering of heterogeneous multi-GPU in UE4 scene can be realized by explicitly calling graphics API, instead of relying on hardware limitations or drivers. Dual GPUs can bring better performance improvement. In this paper, we mainly design a novel multi- thread rendering framework in UE4 scene, which has a better improvement in performance. Under the current multi-GPU implementation, we have expanded the multithreading architecture and added a new RHI Thread, which we call MRHI Thread, to handle the rendering instruction operation for the newly added GPU. Then, in the design of rendering architecture, rendering objects are divided into U Object main thread logic component object, F Render Object rendering thread object, and FRHI Render Object RHI thread object. Moreover, Because the requirement is to render two scenes, the Scene Capture component is used as the entrance of GPU1 rendering. In the test, we used two Nvida RTX3070, and realized parallel rendering of two GPUs. The performance improved by 150-160%, and the parallel efficiency was close to 95%. Under our proposed framework, the parallel efficiency of the GPU will be greatly improved compared with the previous version. [ABSTRACT FROM AUTHOR]

Abstract With the increasing of computing ability, large-scale simulations have been generating massive amounts of data in aerodynamics. Sort-last parallel rendering is the most classical image compositing method for large-scale scientific visualization. However, in the stage of image compositing, the sort-last method may suffer from scalability problem on large-scale processors. Existing image compositing algorithms tend to perform well in certain situations. For instance, Direct Send is well on small and medium scale; Radix-k gets well performance only when the k-value is appropriate and so on. In this paper, we propose a novel method named mSwap for scientific visualization in aerodynamics, which uses the best scale of processors to make sure its performance at the best. mSwap groups the processors that we can use with a (m,k) table, which records the best combination of m (the number of processors in subgroup of each group) and k (the number of processors in each group). Then in each group, using a m-ary tree to composite the image for reducing the communication of processors. Finally, the image is composited between different groups to generate the final image. The performance and scalability of our mSwap method is demonstrated through experiments with thousands of processors.

The scene-rendering mechanism based on binocular vision is one of the key techniques for the VR globe to achieve immersion-type visualization of global 3D scenes. However, this special rendering mechanism also requires that the 3D scene is continuously drawn twice within one frame, which significantly affects the rendering efficiency of VR globes. Therefore, we propose a binocular parallel rendering method. This method first improves the current rendering process of VR globes by assigning the rendering tasks for the left and right camera of VR to be processed on different CPU cores, thereby achieving parallel rendering of binocular scenes. Second, due to the problem of inconsistent resolution of binocular scenes caused by different viewpoints for the left and right cameras, we propose a resolution synchronize algorithm. this algorithm conducts real-time synchronization on the resolution of scene in the rendering process and thus avoids the problem of erroneous binocular stereo matching. Finally, we validate the effectiveness of the method in this paper through experiments. The results of experiments indicate that while the method in this paper can ensure the consistency of binocular scene resolution, it can decrease the frame time of VR globes by approximately 27% on average.

With the increasing of computing ability, large-scale simulations have been generating massive amounts of data in aerodynamics. Sort-last parallel rendering is the most classical image compositing method for large-scale scientific visualization. However, in the stage of image compositing, the sort-last method may suffer from scalability problem on large-scale processors. Existing image compositing algorithms tend to perform well in certain situations. For instance, Direct Send is well on small and medium scale; Radix-k gets well performance only when the k-value is appropriate and so on. In this paper, we propose a novel method named mSwap for scientific visualization in aerodynamics, which uses the best scale of processors to make sure its performance at the best. mSwap groups the processors that we can use with a (m,k) table, which records the best combination of m (the number of processors in subgroup of each group) and k (the number of processors in each group). Then in each group, using a m-ary tree to composite the image for reducing the communication of processors. Finally, the image is composited between different groups to generate the final image. The performance and scalability of our mSwap method is demonstrated through experiments with thousands of processors. [ABSTRACT FROM AUTHOR]

The U.S. Department of Energy's (DOE) Office of Environmental Management (DOE/EM) currently supports an effort to understand and predict the fate of nuclear contaminants and their transport in natural and engineered systems. Geologists, hydrologists, physicists and computer scientists are working together to create models of existing nuclear waste sites, to simulate their behavior and to extrapolate it into the future. We use visualization as an integral part in each step of this process. In the first step, visualization is used to verify model setup and to estimate critical parameters. High-performance computing simulations of contaminant transport produces massive amounts of data, which is then analyzed using visualization software specifically designed for parallel processing of large amounts of structured and unstructured data. Finally, simulation results are validated by comparing simulation results to measured current and historical field data. We describe in this article how visual analysis is used as an integral part of the decision-making process in the planning of ongoing and future treatment options for the contaminated nuclear waste sites. Lessons learned from visually analyzing our large-scale simulation runs will also have an impact on deciding on treatment measures for other contaminated sites.

The scene-rendering mechanism based on binocular vision is one of the key techniques for the VR globe to achieve immersion-type visualization of global 3D scenes. However, this special rendering mechanism also requires that the 3D scene is continuously drawn twice within one frame, which significantly affects the rendering efficiency of VR globes. Therefore, we propose a binocular parallel rendering method. This method first improves the current rendering process of VR globes by assigning the rendering tasks for the left and right camera of VR to be processed on different CPU cores, thereby achieving parallel rendering of binocular scenes. Second, due to the problem of inconsistent resolution of binocular scenes caused by different viewpoints for the left and right cameras, we propose a resolution synchronize algorithm. this algorithm conducts real-time synchronization on the resolution of scene in the rendering process and thus avoids the problem of erroneous binocular stereo matching. Finally, we validate the effectiveness of the method in this paper through experiments. The results of experiments indicate that while the method in this paper can ensure the consistency of binocular scene resolution, it can decrease the frame time of VR globes by approximately 27% on average. [ABSTRACT FROM AUTHOR]

Developing complex, real world graphics applications which leverage multiple GPUs and computers for interactive 3D rendering tasks is a complex task. It requires expertise in distributed systems and parallel rendering in addition to the application domain itself. We present a mature parallel rendering framework which provides a large set of features, algorithms and system integration for a wide range of real-world research and industry applications. Using the $\mathsf{Equalizer}$ Equalizer parallel rendering framework, we show how a wide set of generic algorithms can be integrated in the framework to help application scalability and development in many different domains, highlighting how concrete applications benefit from the diverse aspects and use cases of $\mathsf{Equalizer}$ Equalizer . We present novel parallel rendering algorithms, powerful abstractions for large visualization setups and virtual reality, as well as new experimental results for parallel rendering and data distribution. [ABSTRACT FROM AUTHOR]

Adaptive Mesh Refinement (AMR) is a technique used in numerical simulations to automatically refine (or re-define) certain regions of the physical domain in an infinite difference calculation. AMR data consists of nested hierarchies of data grids. As AMR visualization is still a relatively unexplored topic, our work is motivated by the need to perform effcient visualization of large AMR data sets. We present a software algorithm for parallel direct volume rendering of AMR data using a cell-projection technique on several different parallel platforms. Our algorithm can use one of several different distribution methods, and we present performance results for each of these alternative approaches. By partitioning an AMR data set into blocks of constant resolution and estimating rendering costs of individual blocks using an application specific benchmark, it is possible to achieve even load balancing.

Parallel volume rendering offers a feasible solution to the large data visualization problem by distributing both the data and rendering calculations among multiple computers connected by a network. In sort-last parallel volume rendering, each processor generates an image of its assigned subvolume, which is blended together with other images to derive the final image. Improving the efficiency of this compositing step, which requires interprocesssor communication, is the key to scalable, interactive rendering. The recent trend of using hardware-accelerated volume rendering demands further acceleration of the image compositing step. This paper presents a new optimized parallel image compositing algorithm and its performance on a PC cluster. Our test results show that this new algorithm offers significant savings over previous algorithms in both communication and compositing costs. On a 64-node PC cluster with a 100BaseT network interconnect, we can achieve interactive rendering rates for images at resolutions up to 1024x1024 pixels at several frames per second.

This paper presents a parallel adaptive rendering algorithm and its performance for visualizing time-varying unstructured volume data generated from large-scale earthquake simulations. The objective is to visualize 3D seismic wave propagation generated from a 0.5 Hz simulation of the Northridge earthquake, which is the highest resolution volume visualization of an earthquake simulation performed to date. This scalable high-fidelity visualization solution we provide to the scientists allows them to explore in the temporal, spatial, and visualization domain of their data at high resolution. This new high resolution explorability, likely not presently available to most computational science groups, will help lead to many new insights. The performance study we have conducted on a massively parallel computer operated at the Pittsburgh Supercomputing Center helps direct our design of a simulation-time visualization strategy for the higher-resolution, 1Hz and 2 Hz, simulations.

Adaptive mesh refinement (AMR) is a technique used in numerical simulations to automatically refine (or de-refine) certain regions of the physical domain in a finite difference calculation. AMR data consists of nested hierarchies of data grids. As AMR visualization is still a relatively unexplored topic, our work is motivated by the need to perform efficient visualization of large AMR data sets. We present a software algorithm for parallel direct volume rendering of AMR data using a cell-projection technique on several different parallel platforms. Our algorithm can use one of several different distribution methods, and we present performance results for each of these alternative approaches. By partitioning an AMR data set into blocks of constant resolution and estimating rendering costs of individual blocks using an application specific benchmark, it is possible to achieve even load balancing.

Adaptive mesh refinement (AMR) is a technique used in numerical simulations to automatically refine (or de-refine) certain regions of the physical domain in a finite difference calculation. AMR data consists of nested hierarchies of data grids. As AMR visualization is still a relatively unexplored topic, our work is motivated by the need to perform efficient visualization of large AMR data sets. We present a software algorithm for parallel direct volume rendering of AMR data using a cell-projection technique on several different parallel platforms. Our algorithm can use one of several different distribution methods, and we present performance results for each of these alternative approaches. By partitioning an AMR data set into blocks of constant resolution and estimating rendering costs of individual blocks using an application specific benchmark, it is possible to achieve even load balancing.

Non-photorealistic rendering can be used to illustrate subtle spatial relationships that might not be visible with more realistic rendering techniques. We present a parallel hardware-accelerated rendering technique, making extensive use of multi-texturing and paletted textures, for the interactive non-photorealistic visualization of scalar volume data. With this technique, we can render a 512x512x512 volume using non-photorealistic techniques that include tone-shading, silhouettes, gradient-based enhancement, and color depth cueing, as shown in the images on the color plate, at multiple frames second. The interactivity we achieve with our method allows for the exploration of a large visualization parameter space for the creation of effective illustrations. Keywords: interactive visualization, non-photorealistic rendering, scientific visualization, silhouette, texture graphics hardware, visual perception, volume rendering, parallel rendering

We present a scalable volume rendering technique that exploits lossy compression and low-cost commodity hardware to permit highly interactive exploration of time-varying scalar volume data. A palette-based decoding technique and an adaptive bit allocation scheme are developed to fully utilize the texturing capability of a commodity 3-D graphics card. Using a single PC equipped with a modest amount of memory, a texture capable graphics card, and an inexpensive disk array, we are able to render hundreds of time steps of regularly gridded volume data (up to 42 millions voxels each time step) at interactive rates. By clustering multiple PCs together we demonstrate the data-size scalability of our method. The frame rates achieved make possible the interactive exploration of data in the temporal,spatial, and transfer function domains. A comprehensive evaluation of our method based on experimental studies using data sets (up to 134 millions voxels per time step) from turbulence flow simulations is also presented. Keywords--Compression, disk I/O, high performance computing, out-ofcore processing, parallel rendering, PC, scalable algorithms, scientific visualization, texture hardware, time-varying data, transform encoding, volume rendering

Our ability to study and understand complex, transient phenomena is critical to solving many scientific and engineering problems. This paper addresses the issues in storage space, I/O, and rendering for visualizing time-varying volume data. A suite of effective techniques, including encoding, parallel rendering, and runtime visualization techniques are discussed. In particular, we introduce a new hardware-assisted technique coupled with a compression scheme based on Discrete Cosine Transform which allows interactive exploration in both spatial and temporal domains of the data using a PC with a commodity graphics card.

To seek a low-cost, extensible solution for the large-scale data visualization problem, a visual computing system is designed as a result of a collaboration between industry and government research laboratories in Japan, also with participation by researchers in U.S. This scalable system is a commodity PC cluster equipped with the VolumePro 500 volume graphics cards and a specially designed image compositing hardware. Our performance study shows such a system is capable of interactive rendering 5123 and 10243 volume data and highly scalable. In particular, with such a system, simulation and visualization can be performed concurrently which allows scientists to monitor and tune their simulations on the fly. In this paper, both the system and hardware designs are presented.

This paper presents a work-in-progress. The objective of our study is to derive an optimal design for high-performance rendering of irregular-grid volume data on the increasingly popular, distributed shared-memory parallel supercomputers. We experiment with a multi-threaded volume rendering algorithm for three-dimensional unstructured-grid data and discuss its performance on the SGI Origin 2000. Our preliminary test results demonstrate good rendering rates with a moderate number of processors. But we observe surprisingly poor processor scalability. We thus investigate how task partitioning and runtime memory management affect, in particular, the scalability of parallel rendering. We discover that in terms of algorithmic complexity and programming effort to attain the optimal performance on a shared memory architecture it can be as challenging and demanding as on a distributed-memory architecture.

This paper reports the performance of a parallel volume rendering algorithm for visualizing a large-scale unstructured-grid dataset produced by a three dimensional aerodynamics simulation. This dataset, containing over 18 million tetrahedra, allows us to extend our performance results to a problem which is more than 30 times larger than the one we examined previously. This high resolution dataset also allows us to see fine, three-dimensional features in the flow field. All our tests were performed on the SGI/Cray T3E operated by NASA's Goddard Space Flight Center. Using 511 processors, a rendering rate of almost 9 million tetrahedra/second was achieved with a parallel overhead of 26%.

Rendering textures in real-time environments is a key task in computer graphics. This paper presents a new parallel patch-based method which allows repeatable sampling without cache, and does not create visual repetitions. Interchangeable patches of arbitrary shape are prepared in a preprocessing step, such that patches may lie over the boundary of other patches in a repeating tile. This compresses the example texture into an infinite texture map with small memory requirements, suitable for GPU and ray-tracing applications. The quality of textures rendered with this method can be tuned in the offline preprocessing step, and they can then be rendered in times comparable to Wang tiles. Experimental results demonstrate combined benefits in speed, memory requirements, and quality of randomisation when compared to previous methods. [ABSTRACT FROM AUTHOR]

This paper presents a parallel visualization technique for illustrative rendering of dense three-dimensional (3D) geometry data sets. Our approach maps the depth information in each geometry onto various visual dimensions of graphical representations, including shape, color, brightness, transparency, and size, to achieve legible display in dense geometry environments where visual clutters often hinder perception and navigation in the visualizations. At the same time, we leverage legacy CPU computing power to overcome performance challenges as a result of the depth-dependent illustrations used for the visual legibility enhancement. This is realized by a novel parallel rendering algorithm we developed particularly for illustrative visualizations of depth-dependent stylized dense geometries at interactive frame rates. While the computation could be performed atop modern GPU devices, we target a parallel visualization framework that enables it to efficiently run on commodity CPUs, which are much more available than GPUs for ordinary users. We evaluated our framework with visualizations of depth-stylized geometries derived from 3D diffusion tensor MRI data, by comparing its efficiency with several other alternative parallelization platforms with respect to the same computations. Results show that our approach can efficiently render highly dense 3D geometry data sets and, thus, it offers not only an alternative and complementary, but also more adoptable, solution to users in contrast to parallel visualization environments that rely on GPUs. [ABSTRACT FROM AUTHOR]