Sidd puri microsoft

We'll do our best to fix them. Check all that apply - Please note that only the first page is available if you have not selected a reading option after clicking "Read Article". Include any more information that will help us locate the issue and fix it faster for you. Programmers can either learn graphics APIs and convert their applications to use graphics pipeline operations or they can use stream programming abstractions of GPUs.

We describe Accelerator, a system that uses data parallelism to program GPUs for general-purpose uses instead. Programmers use a conventional imperative programming language and a library that provides only high-level data-parallel operations. No aspects of GPUs are exposed to programmers. We describe the compilation techniques used to do this. We evaluate the effectiveness of using data parallelism to program GPUs by providing results for a set of compute-intensive benchmarks.

We compare the performance of Accelerator versions of the benchmarks against hand-written pixel shaders. Continue with Facebook. How was the reading experience on this article? The text was blurry Page doesn't load Other:. Details Include any more information that will help us locate the issue and fix it faster for you. Thank you for submitting a report! Submitting a report will send us an email through our customer support system. Submit report Close. ISBN doi Recommended Articles Loading There are no references for this article.

Read and print from thousands of top scholarly journals. Already have an account? Keyphrases document processing high performance convolutional neural network gpu implementation unrolled convolution matrix-matrix product representation character recognition problem state-of-the-art recognizers convolutional neural network matrix-matrix product convolutional layer novel approach pixel shader traditional multi-layer perceptrons matrix product algebra subroutine unrolling convolution.

Powered by:. Note : im2col does not improve the time complexity of Convolution. If channel is not the inner most layer, we need to transform it to the above format. Alternatives to im2col Alternatives to im2col are: im2row kn2col kn2row In im2row, spatial locality is better compared to im2col and hence, im2row performs better than im2col on CPU.