high performance computing on graphics processing units: hgpu.org

We present a novel dynamic configuration technique for deep neural networks that permits step-wise energy-accuracy trade-offs during runtime. Our configuration technique adjusts the number of channels in the network dynamically depending on response time, power, and accuracy targets. To enable this dynamic configuration technique, we co-design a new training algorithm, where the network is incrementally […]

This paper illustrates how GPU computing can be used to accelerate computational fluid dynamics (CFD) simulations. For sparse linear systems arising from finite volume discretization, we evaluate and optimize the performance of Conjugate Gradient (CG) routines designed for manycore accelerators and compare against an industrial CPU-based implementation. We also investigate how the recent advances in […]

In this article, we propose a deep neural network (DNN) architecture called Input Output Deep Architecture (IODA) for solving the problem of image labeling. IODA directly links a whole image to a whole label map, assigning a label to each pixel using a single neural network forward step. Instead of designing a handcrafted a priori […]

In this paper, the suitability of implementing parallel homomorphic word searching on Intel Xeon Phi coprocessors is evaluated for the first time. Homomorphic encryption allows to produce a cryptogram that encrypts the result of applying some values to any function, even when the input values are encrypted and without access to the privatekey. For example, […]

Advances are reported in the three pillars of computational science achieving a new capability for understanding dynamic plasma phenomena outside of local thermodynamic equilibrium. A continuum kinetic model for plasma based on the Vlasov-Maxwell system for multiple particle species is developed. Consideration is added for boundary conditions in a truncated velocity domain and supporting wall […]

Graphical Processing Units are de-facto standard for acceleration of data parallel tasks in high performance computing. They are widely used to accelerate batch machine learning algorithms. High-end discrete GPUs are characterized by a very high number of cores (thousands), high bandwidth memory optimized for the stream access and high power requirements. Integrated GPUs are characterized […]

Paper publication All paper submissions will be peer reviewed and evaluated based on originality, research content, relevance to conference, contributions, and readability. All accepted papers will be published in one of the indexed Journals after proper registration and presentation. – Journal of Computers(JCP, ISSN: 1796-203X) Indexed by: ULRICH’s Periodicals Directory; Google Scholar; INSPEC; etc. – […]

Paper Publication: All papers are reviewed using a single-blind review process, the accepted paper will be published in journal. Journal of Communications (JCM, ISSN: 1796-2021); Journal of Software(JSW, ISSN: 1796-217X); Journal of Computers(JCP, ISSN: 1796-203X); International Journal of Future Computer and Communication (IJFCC, ISSN: 2010-3751); International Journal of Computer Theory and Engineering (IJCTE, ISSN: 1793-8201); […]

Paper Publication Accepted papers of ICOSP 2016 could be published in: International Journal of Signal Processing Systems (IJSPS)

Publication: All accepted papers will be published in the volume of MATEC Web of Conferences (ISSN: 2261-236X), being indexed by Ei Compendex, Inspec, DOAJ, CPCI (Web of Science) and Scopus. Submission Methods: Full Paper (publication and oral presentation) Abstract (oral presentation only) Electronic Submission System (.pdf) https://www.easychair.org/conferences/?conf=icmes2016

Singular Value QR (SVQR) can orthonormalize a set of dense vectors with the minimum communication (one global reduction between the parallel processing units, and BLAS-3 to perform most of its local computation). As a result, compared to other orthogonalization schemes, SVQR obtains superior performance on many of the current computers, where the communication has become […]

A fifth-order implicit Runge-Kutta method and two fourth-order exponential integration methods equipped with Krylov subspace approximations were implemented for the GPU and paired with the analytical chemical kinetic Jacobian software pyJac. The performance of each algorithm was evaluated by integrating thermochemical state data sampled from stochastic partially stirred reactor simulations and compared with the commonly […]