Tag Archives: gpu computing

CUDA 2.0 Available

CUDA 2.0 is available here: CUDA Zone. To take advantage of CUDA, you need Forceware 177.84 or better (177.89). Samples work with GeForce 8/9 series.

NVIDIA CUDA technology is the world’s only C language environment that enables programmers and developers to write software to solve complex computational problems in a fraction of the time by tapping into the many-core parallel processing power of GPUs.
—-
[via]

CUDA: GPU Usage and Data Structure Design

Here is a thesis that discusses the usage of NVIDIA’s CUDA in two applications:
– Einstein@Home: a distributed computing software
– OpenSteer: a game-like application.

CUDA exposes the GPU processing power in the C programming language and can be integrated in existing applications with ease. But in order to exploit the power a GPU can deliver, one has to design the data structures in order to become optimized for CUDA.

Download the thesis here: GPU usage and data structure design (1556)

-source-

NVIDIA CUDA: One Year After

The hardware.fr staff has been invited by NVIDIA to make a point on CUDA. Since this exellent article is written in french, I’ll try to highlight the interesting parts.

One of the new thing in CUDA 2.0 is, according to hardware.fr, the adding in the CUDA compiler of an optimzed profile for multicores x86 CPUs. Currently, CUDA code is splitted in two parts: one part processed by the CPU and the other one by the GPU via the CUDA compiler.
The new thing is that we can now compile the GPU code explicitly for the CPU in order to take advantage of multicores capabilities of the latest CPUs.

Another new thing is Tesla Series 10. NVIDIA has equiped all Tesla 10 products with 4Gb of graphics memory by GPU (recall that GeForce GTX 280 has 1Gb of memory). This boost in memory amount is useful in situations where dataset to be processed are very large.


A Tesla 10 card has only 6-pin PCI-Express power connector (the 8-pin is optional – a GeForce GTX 280 has one 6-pin and one 8-pin an both are required!). The reason is in GPU Computing the GPU has a lower power consumption because some transitors dedicated to 3D graphics are not used.

The article shows also some practical cases where CUDA is used: financial analysis, medical imagery (3D scans) and password recovering.


Read the complete article HERE – in french only
[via]

Real-time Visual Tracker by Stream Processing

This work describes the implementation of a real-time visual tracker that targets the position and 3D pose of objects (specifically faces) in video sequences. The use of GPUs for the computation and efficient sparse-template-based particle filtering allows real-time processing even when tracking multiple faces simultaneously in high-resolution video frames. Using a GPU and the NVIDIA CUDA technology, performance improvements as large as ten times compared to a similar CPU-only tracker are achieved.

Real-time Visual Tracker by Stream Processing by Oscar Mateo Lozano, and Kazuhiro Otsuka. Journal of Signal Processing Systems.


-source-
Thanks to tho for the news.

Tesla 10 and CUDA 2.0: Technical Analysis and Performance

CUDA was announced along with G80 in November 2006, released as a public beta in February 2007, and then finally hit the Version 1.0 milestone in June 2007 along with the launch of the G80-based Tesla solutions for the HPC market. Today, Beyond3D looks at the next stage in the CUDA/Tesla journey: GT200-based solutions, CUDA 2.0, and the overall state of NVIDIA’s HPC business.

Read the article HERE.

AMD refines its approach to Stream Computing

Just like Nvidia, AMD provides developers with a high-level application programming interface (API) to tap into its latest graphics processors for non-graphics compute applications. Unlike Nvidia, though, AMD doesn’t make very much noise about what it’s doing in that area. Curious, the guys at The Tech Report got on the phone with AMD Stream Computing Director Patti Harrell and asked her to shed a little light on AMD’s Stream Computing initiative.

Read the complete article HERE.


-source-

Making CPU and GPU play nice together

Do you know what CUDA and OpenCL stand for and how they could make your computer 50 times faster? If so, you can safely jump to the “Ending the mess” section below. Otherwise read on for a gentle introduction.

A computer has two important processing units: the CPU and GPU. Think of them as the two brothers in Rain Man. The GPU is the ultimate autistic savant. He’s really, really good at counting stuff and doing a lot of complex math at the same time.

The CPU is your regular guy. He can do all kinds of stuff that the savant can’t. He goes along well with everybody, as long as they speak English. If he learns to take advantage of the savant, the two of them can do amazing things like count cards at Poker.

In other words, the GPU is natural at some operations that involve repetitive calculations, like those necessary for drawing 3D graphics and doing basic image manipulation.

Read the rest of this article HERE.

NVIDIA GPU Compute FAQ

Benchmark Reviews proposes an article about GPU computing with CUDA and GeForce based graphics cards.

Read the complete GPU Compute FAQ HERE.

Terms such as “heterogeneous computing” and “parallel computing” are going to be used as often as the term “video card” is used in a product review. You won’t want to miss this evolution in graphics technology, because we are witness to a pivitol moment in time when computers are going to stop being filled with familiar single-purpose hardware. Benchmark Reviews offers this FAQ to help our readers understand what is happening, and help introduce them to what is coming. We don’t want anyone to be left in the cold when the rest of the world learns how the GPU is learning to be a CPU.