Compute Unified Device Architecture Application Suitability

Hwu, Wen-mei W.; Rodrigues, Christopher; Ryoo, Shane; Stratton, John A.

doi:10.1109/mcse.2009.48

Cited by 46 publications

(14 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Programming systems such as CUDA [20], ATI Stream Technology [1], and OpenCL [13] allow programmers to parallelize an application into thousands of threads each of which executes the same code. Previous work [23,9] has shown that some applications experience an order of magnitude speedup when run on a GPU instead of a CPU. GPUs Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page.…”

Section: Introductionmentioning

confidence: 99%

Improving GPU performance via large warps and two-level warp scheduling

Narasiman

Shebanow

Lee

et al. 2011

Proceedings of the 44th Annual IEEE/ACM International Symposium on Microarchitecture

352

246

View full text Add to dashboard Cite

Due to their massive computational power, graphics processing units (GPUs) have become a popular platform for executing general purpose parallel applications. GPU programming models allow the programmer to create thousands of threads, each executing the same computing kernel. GPUs exploit this parallelism in two ways. First, threads are grouped into fixed-size SIMD batches known as warps, and second, many such warps are concurrently executed on a single GPU core. Despite these techniques, the computational resources on GPU cores are still underutilized, resulting in performance far short of what could be delivered. Two reasons for this are conditional branch instructions and stalls due to long latency operations.To improve GPU performance, computational resources must be more effectively utilized. To accomplish this, we propose two independent ideas: the large warp microarchitecture and two-level warp scheduling. We show that when combined, our mechanisms improve performance by 19.1% over traditional GPU cores for a wide variety of general purpose parallel applications that heretofore have not been able to fully exploit the available resources of the GPU chip.

show abstract

Section: Introductionmentioning

confidence: 99%

Improving GPU performance via large warps and two-level warp scheduling

Narasiman

Shebanow

Lee

et al. 2011

Proceedings of the 44th Annual IEEE/ACM International Symposium on Microarchitecture

352

246

View full text Add to dashboard Cite

show abstract

“…All computationally demanding parts of the state estimator are implemented in CUDA [16] in order to achieve the computational speed required for online control of the robot. The particle filter was setup to run with 3200 particles, with the filter running at 80 Hz.…”

Section: A Experimental Setup and Camera Calibrationmentioning

confidence: 99%

Particle filter based tracking of free-swinging objects for visual servoing

Myhre

Transeth

Egeland

2014

2014 IEEE International Conference on Robotics and Biomimetics (ROBIO 2014)

View full text Add to dashboard Cite

We present a visual servoing system for the synchronization of a robot with a dynamic rigid body using a particle filter-based state estimator. We experimentally demonstrate its application for robotic loading of parts onto oscillating hangers in an industrial paint line. The particle filter is implemented with a physical dynamical model of the hanger, using the Euler equations for a pendulum with three degrees of freedom. The visual servoing system is successfully demonstrated using two cameras to get a good 3D estimate of the hanger motion. The experimental results indicate a maximum tracking error of 0.1 degrees.

show abstract

“…Hence we believe that there is a real contribution to be made in this area, specially now, when GPU Computing has taken such a huge step forward, with the appearance of tools such as NVIDIA's CUDA architecture [20], which will be summarised in the following section.…”

Section: Related Workmentioning

confidence: 99%

“…We will demonstrate that the SIMD features of GPUs provide a means of dealing with the scalability of highly parallelisable algorithms operating on large data structures, drastically improving overall processing rates comparing with CPU implementations, therefore allowing real-time performance. Moreover, we will demonstrate that CUDA is a very useful tool for GPU programming, as it provides a simplified yet powerful abstraction to graphic card intricacies [20].…”

Section: Introductionmentioning

confidence: 99%

Bayesian real-time perception algorithms on GPU

Ferreira

Lobo

Dias

2010

J Real-Time Image Proc

View full text Add to dashboard Cite

In this text we present the real-time implementation of a Bayesian framework for robotic multisensory perception on a graphics processing unit (GPU) using the Compute Unified Device Architecture (CUDA). As an additional objective, we intend to show the benefits of parallel computing for similar problems (i.e. probabilistic grid-based frameworks), and the user-friendly nature of CUDA as a programming tool. Inspired by the study of biological systems, several Bayesian inference algorithms for artificial perception have been proposed. Their high computational cost has been a prohibitory factor for realtime implementations. However in some cases the bottleneck is in the large data structures involved, rather than the Bayesian inference per se. We will demonstrate that the SIMD (single-instruction, multiple-data) features of GPUs provide a means for taking a complicated framework of relatively simple and highly parallelisable algorithms operating on large data structures, which might take up to several minutes of execution with a regular CPU implementation, and arrive at an implementation that executes in the order of tenths of a second. The implemented multimodal perception module (including stereovision, binaural sensing and inertial sensing) builds an egocentric representation of occupancy and local motion, the Bayesian Volumetric Map (BVM), based on which gaze shift decisions are made to perform active exploration and reduce the entropy of the BVM. Experimental results show that the real-time implementation successfully drives the robotic system to explore areas of the environment mapped with high uncertainty.

show abstract

Compute Unified Device Architecture Application Suitability

Cited by 46 publications

References 6 publications

Improving GPU performance via large warps and two-level warp scheduling

Improving GPU performance via large warps and two-level warp scheduling

Particle filter based tracking of free-swinging objects for visual servoing

Bayesian real-time perception algorithms on GPU

Contact Info

Product

Resources

About