New SDK ups GPU-CPU data transfer rate

Move your data between CPU and GPU at up to 15 GB per second

AMD (NYSE:AMD) has just released version 2.5 of its AMD Accelerated Parallel Processing (APP) Software Development Kit, according to a press release. Updated

The SDK builds on OpenCL that was invented by Apple but that has now also been embraced by both Intel and AMD and which is now available as open source. According to AMD the latest version of the SDK fully supports the A-series APUs with integrated GPU.

The SDK allows for fast transfer of data between the CP and the GPU part of the APU. For the A-series the speeds are up to 15 GB per second.

“Improving performance and programmability on AMD platforms is a key initiative for us, and we work closely with developers to help us make the APU the best possible development platform,” said Manju Hegde, corporate vice president, AMD Fusion Experience Program. “With these latest SDK refinements, we are supporting continued growth of the ecosystem and building on the momentum generated by the successful AMD Fusion Developer Summit held in June of this year.”

Other improved features of the SDK include enhancements to the OpenCL runtime designed to enable more efficient use of available GPU compute capabilities by reducing kernel launch times and PCI Express transfer overheads.

In addition the SDK also supports the utilization of discrete GPU resources that enables scaling across multiple GPUs.S|A

Updated: Invention of OpenCL corrected.

The following two tabs change content below.