SYSTEMS
GPU Systems releases Libra 3.0
- Heterogeneous Cloud Computing SDK available across PC, Tablet and Mobile Devices including a new virtualizing function for cloud services
- local and remote CPU – GPU computing access - directly accessible via C/C++, Java, C# and matlab
- Enabling accelerated mobile and tablet computing through the combination of CPU and GPU technologies across major Mobile, Tablet operating systems, Windows RT, iOS and Android accessed remotely or locally via standard highlevel programming environments C/C++, Java, C#, matlab etc.
At GPU Systems headquarters, the company has released a new version of the Libra Platform & SDK with enhanced support for developers to uniformly accelerate software applications across major Mobile and Tablet operating systems, Windows RT, iOS and Android. Libra Compute API for software developers offers performance portability and direct compute access remotely or locally via standard programming environments C/C++, Java, C# and matlab, etc.
"With this release we show that the very latest advanced compute technologies are available to develop massively accelerated software applications on Mobile and Tablet devices making full use of Heterogeneous computing via standard programming languages & environments. It is now possible to create accelerated software to execute locally and remotely on a broad range of CPU and GPU devices and across major Mobile, Tablet and PC platforms without any intervention or modification to program code," said GPU Systems founder Marco Hjerpe.
Key Facts
* Unleash the Teraflop performance of CPUs and GPUs and future compute accelerators.
* Libra API direct compute access via standard programming languages, C/C++, Java, C#, matlab etc - Major operating systems supported, Win, Mac, Linux, WinRT, iOS and Android.
* Create accelerated software to execute math operations locally and remotely on devices.
* Develop code once, deploy on numerous CPU and GPU devices across Mobile, Tablet and PC platforms.
* More performance with less code compared to hardware specific lowlevel APIs, easier to maintain applications & algorithms and faster time to market, no need to learn new tools or programming languages.
For more inofrmation, visit: http://www.gpusystems.com .