Removing Bottlenecks with Python; Programming Models for HSA, See you at SC15 + a bit more


Programming Models for Heterogeneous Systems

An ideal programming model would be one that is highly productive and consists of a rich set of features for performant programming. Is there a best one? Check out the analysis in this blog.

ACL Beta 2 Brings a Slew of Features & Improvements

The ACL 1.0 Beta 2 release is now available and it includes a ton of interesting goodies, like: AutoGemm, the new high-performing GEneric Matrix Matrix multiplication (GEMM) backend for clBLAS. This is a suite of Python scripts that provide a host of benefits. ACL Beta 2 also includes a new single precision SpM-SpM (SpGEMM) function, an improved version of clSPARSE, and, yes, lots more. Get the details in this new blog post.

Webinar: Blowing the Doors off Bottlenecks with Python on AMD APUs

Learn how to speed up your Python programs using the integrated GPU on AMD APUs. Learn more about this new webinar and sign up now. There is no cost for this webinar but seating is limited.

clFFT Pre-callback: A Faster Way to Pre-Process Data

The math library group at AMD is continuously looking for areas of improvement and working towards optimizing the same. This blog explains the speedup improvements in clFFT 2.6.1 over previous versions, as well as competitive performance against, well, the competitor. How does it stack up?

Forcing High Performance on the GPU

AMD Enduro™ technology regulates your notebook’s GPU, giving you an instant boost in graphics performance when you need it and consuming almost zero watts of power when you don’t. That can give you great battery life, but what do you do, as a developer, when you want to force your application to run on the discrete GPU? Find out how in this latest blog post.

Can you improve programmability with Fine-Grained SVM?

The AMD APP SDK 3.0 final release is now available, and includes some new examples that were not present in the beta release. This blog takes a look at two of those examples and several features that improve programmability and performance.

SC15 Returns to Austin

The International Conference for High Performance Computing, Networking, Storage and Analysis will be in Austin for its 27™ annual conference this year, and so will we. Stop by and see AMD at booth 727. We’ll have a few surprises we think you’ll like.

High Performance Python on AMD APUs

Speed up compute intensive analytics 2-5X with AMD APUs and Anaconda, a modern open source analytics platform powered Python, that includes Numba a Python compiler.


AMD Asynchronous Shaders Demo

Leverage the Speed of OpenCL™ with AMD Math Libraries

ACMLScript: Adjusting Heuristics (Part 3 of 3)


Notify of
Inline Feedbacks
View all comments