GPGPU » Category: Business :: GPGPU.org

NVIDIA Kepler GK110 Architecture White Paper

Mark Harris — Mon, 21 May 2012 00:35:21 +0000

NVIDIA Kepler GK110 Die Shot

This white paper describes the new Kepler GK110 Architecture from NVIDIA.

Comprising 7.1 billion transistors, Kepler GK110 is not only the fastest, but also the most architecturally complex microprocessor ever built. Adding many new innovative features focused on compute performance, GK110 was designed to be a parallel processing powerhouse for Tesla® and the HPC market.

Kepler GK110 will provide over 1 TFlop of double precision throughput with greater than 80% DGEMM efficiency versus 60‐65% on the prior Fermi architecture.

In addition to greatly improved performance, the Kepler architecture offers a huge leap forward in power efficiency, delivering up to 3x the performance per watt of Fermi.

The paper describes features of the Kepler GK110 architecture, including

Dynamic Parallelism;
Hyper-Q;
Grid Management Unit;
NVIDIA GPUDirect™;
New SHFL instruction and atomic instruction enhancements;
New read-only data cache previously only accessible to texture;
Bindless Textures;
and much more.

CUVILib v1.2 released

dom — Thu, 17 May 2012 07:21:17 +0000

TunaCode has released CUVILib v1.2, a library to accelerate imaging and computer vision applications. CUVILib adds acceleration to Imaging applications from Medical, Industrial and Defense domains. It delivers very high performance and supports both CUDA and OpenCL. Modules include color operations (demosaic, conversions, correction etc), linear/non-linear filtering, feature extraction & tracking, motion estimation, image transforms and image statistics.

More information, including a free trial version: http://www.cuvilib.com/

New Libra Platform version released

dom — Sat, 21 Apr 2012 09:05:14 +0000

Libra Platform is a GPGPU-Heterogeneous Compute API and runtime environment available on Windows, Mac and Linux. Libra Compute API offers performance portability and direct compute access via standard programming environments C/C++, Java, C# and Matlab to execute math operations on top of current and future compute architectures, including the latest GPUs, x86/x64 CPUs and with broad support for compute devices compatible with low level specific APIs – OpenCL, CUDA, OpenGL and standard x86/x64 compute APIs.

2 Day CUDA Workshop, May 5-6 2012, Berlin, Germany

dom — Sat, 21 Apr 2012 08:59:29 +0000

A 2 day CUDA workshop is taking place in Berlin, Germany on May 5 and 6 2012. Course details, outline and prices are available at http://cuda.eventbrite.com.

Acceleware OpenCL™ Training in NYC

Mark Harris — Wed, 29 Feb 2012 01:56:31 +0000

Developed in partnership with AMD, this four day course is designed for GPU Programmers who are looking to develop comprehensive skills in writing and optimizing applications that fully leverage the multi-core processing capabilities of the GPU.

Delivered by Acceleware’s Developers, who provide real world experience and examples, the training comprises classroom lectures and hands-on tutorials. Each student will be supplied with a laptop equipped with an AMD Fusion APU for the duration of the course. Small class sizes maximize learning and ensure a personal educational experience.

SpeedIT 2.0 released

dom — Fri, 24 Feb 2012 06:55:07 +0000

SpeedIT 2.0 and the SpeedIT plugin to OpenFOAM have been released. New features include:

One of the fastest Sparse Matrix Vector Multiplication worldwide.
Faster Conjugate Gradient and BiConjugate Gradient solvers.
State-of-the-art CMRS format for storing sparse matrices. The format requires less memory than CRS or HYB (from CUSPARSE and CUSP).
Faster acceleration in OpenFOAM (Computational Fluid Dynamics).

More information is available at http://speed-it.vratis.com.

Performance of SpMV in CUSPARSE, CUSP and SpeedIT

dom — Sat, 14 Jan 2012 12:43:31 +0000

The SpeedIt team recently compared and benchmarked the SpMV performance of CUSPARSE 4.0, CUSP 0.2.0 and SpeedIT 2.0 on 23 randomly chosen matrices from University Florida Matrix Collection. Comparisons were done on a Tesla C2050 in single and double precision. The full report is available at http://wp.me/p1ZihD-1.

Acceleware 4 Day CUDA Course

Mark Harris — Fri, 06 Jan 2012 12:08:40 +0000

Partnering with NVIDIA and Microsoft, this four day course is designed for Programmers who are looking to develop comprehensive skills in writing and optimizing applications that fully leverage the multi-core processing capabilities of the GPU.

Delivered by Acceleware’s Developers, who provide real world experience and examples, the training comprises classroom lectures and hands-on tutorials. Each student will be supplied with a laptop equipped with NVIDIA GPUs for the duration of the course. Small class sizes maximize learning and ensure a personal educational experience.

Introduction to Generic Accelerated Computing with Libra SDK

dom — Wed, 30 Nov 2011 07:35:49 +0000

Libra SDK is a sophisticated runtime including API, sample programs and documentation for massively accelerating software computations. This introduction tutorial provides an overview and usage examples of the powerful Libra API & math libraries executing on x86/x64, OpenCL, OpenGL and CUDA technology. Libra API enables generic and portable CPU/GPU computing within software development without the need to create multiple, specific and optimized code paths to support x86, OpenCL, OpenGL or CUDA devices. Link to PDF: www.gpusystems.com/doc/LibraGenericComputing.pdf

GPU Virtualization for Dynamic GPU Provisioning

dom — Fri, 18 Nov 2011 08:23:53 +0000

From a recent press release:

Taipei, November 18, 2011: Zillians, a leading cloud solution provider specializing in high performance computing, GPU virtualization middleware and massive multi-player online game (MMOG) platforms today announced the availability of vGPU – the world’s first commercial virtualization solution for decoupling GPU hardware from software. Traditionally, physical GPUs must reside on the same machine running GPU code. This severely hampers GPU cloud deployment due to the difficulty of dynamic GPU provisioning. With vGPU technology, bulky hardware is no longer a limiting factor. vGPU introduces a thin, transparent RPC layer between local application and remote GPU, enabling existing GPU software to run without any modification on a remote GPU resource.

Utilizing Mellanox Technologies InfiniBand RDMA I/O solutions and NVDIA GPUDirect technology, Zillians vGPU achieved peak data transfer rates between application and GPU hardware. This opens the door to a broader range of GPU applications that can take advantage of virtualized GPU resources, and also a boon for service providers or cloud operators looking to deploy next generation cloud services based on scalable GPU and Infiniband technologies. As interest in GPU-based HPC clusters continues to grow, vGPU represents a critical enabler for systems to realize resource sharing and improve system utilization across a variety of application domains.

“We are excited to collaborate with Zillians to provide new capabilities for high-performance and massive multi-player online gaming cloud platforms,” said Gilad Shainer, Senior Director, Market Development at Mellanox Technologies. “Mellanox InfiniBand I/O solutions with native RDMA support and GPU acceleration provide the essential technology for Zillians’ innovative vGPU solution.”

Availability

Zillians vGPU is now available for early evaluation request. Version 1.0 featuring support for NVIDIA CUDA 4.0 runtime and driver APIs will be released by the end of 2011. Version 2.0 featuring further support for OpenCL 1.1 is scheduled for release in 2012 Q1. To learn more, please visit www.zillians.com/vgpu.