high performance computing on graphics processing units: hgpu.org

A buffer k-d tree is a k-d tree variant for massively-parallel nearest neighbor search. While providing valuable speed-ups on modern many-core devices in case both a large number of reference and query points are given, buffer k-d trees are limited by the amount of points that can fit on a single device. In this work, […]

We introduce the method of compressive dynamic mode decomposition (cDMD) for robustly performing real-time foreground/background separation in high-definition video. The DMD method provides a regression technique for least-square fitting of video snapshots to a linear dynamical system. The method integrates two of the leading data analysis methods in use today: Fourier transforms and Principal Components. […]

Cache memory, although important for boosting application performance, is also a source of execution time variability, and this makes its use difficult in systems requiring worst case execution time (WCET) guarantees. Cache locking is a promising approach for simplifying WCET estimation and providing predictability and hence, several commercial processors provide ability for locking cache. However, […]

The paper discusses an interest rate term structure decomposition method that breaks from the conventional, in that it does not superimpose any model, form or structure on the decomposition output – hence, the term free-form. The premise is simple: if the model does not presuppose any structure beforehand, and if the structure underlying the input […]

Road lane detection are classical requirements for advanced driving assistant systems. With new computer technologies, lane detection algorithms can be exploited on COTS platforms. This paper investigates the use of OpenCL and develop a particle-filter based lane detection algorithm that can tune the trade-off between detection accuracy and speed. Our algorithm is tested on 14 […]

The precise semantics of floating-point arithmetic programs depends on the execution platform, including the compiler and the target hardware. Platform dependencies are particularly pronounced for arithmetic-intensive parallel numeric programs and infringe on the highly desirable goal of software portability (which is nonetheless promised by heterogeneous computing frameworks like OpenCL): the same program run on the […]

Large-scale simulations can produce hundreds of terabytes to petabytes of data, complicating and limiting the efficiency of work-flows. Traditionally, outputs are stored on the file system and analyzed in post-processing. With the rapidly increasing size and complexity of simulations, this approach faces an uncertain future. Trending techniques consist of performing the analysis in-situ, utilizing the […]

In recent years, the use of Graphics Processing Units (GPUs) for data mining tasks has become popular. With modern processors integrating both CPUs and GPUs, it is also important to consider what tasks benefit from GPU processing and which do not, and apply a heterogeneous processing approach to improve the efficiency where applicable. Similarity search, […]

Modern applications including graphics, multimedia, web search, and data analytics not only can benefit from acceleration, but also exhibit significant degrees of tolerance to imprecise computation. This amenability to approximation provides an opportunity to trade quality of the results for higher performance and better resource utilization. Exploiting this opportunity is particularly important for FPGA accelerators […]

There are numerous way to express parallelism which can make it challenging for developers to verify these programs. Many tools only target a single dialect but the Concurrency Intermediate Verification Language (CIVL) targets MPI, Pthreads, and CUDA. CIVL provides a general concurrency model that can represent pro- grams in a variety of concurrency dialects. CIVL […]

High performance graph analytics are critical for a long list of application domains. In recent years, the rapid advancement of many-core processors, in particular graphical processing units (GPUs), has sparked a broad interest in developing high performance parallel graph programs on these architectures. However, the SIMT architecture used in GPUs places particular constraints on both […]

We show that an end-to-end deep learning approach can be used to recognize either English or Mandarin Chinese speech–two vastly different languages. Because it replaces entire pipelines of hand-engineered components with neural networks, end-to-end learning allows us to handle a diverse variety of speech including noisy environments, accents and different languages. Key to our approach […]