[Beowulf] Mont-Blanc supercomputer powered by Samsung's Exynos

[Eugen Leitl]

http://www.tgdaily.com/hardware-features/67548-mont-blanc-supercomputer-powered-by-samsungs-exynos

Mont-Blanc supercomputer powered by Samsung's Exynos

Posted on November 17, 2012 - 10:00 by Trent Nouveau

The European-based Mont-Blanc project has selected Samsung's flagship Exynos
platform as the building block to power its first integrated low power- High
Performance Computing (HPC) prototype.

According to Alex Ramirez, coordinator of the Mont-Blanc project, the goal of
the initiative is to design a new type of computer architecture capable of
setting future global HPC standards.

As we've previously discussed on TG Daily, the Samsung Exynos 5 Dual is built
on 32nm low-power HKMG (High-K Metal Gate) and features a dual-core 1.7GHz
mobile CPU based on ARM Cortex-A15 architecture paired with an integrated
Mali-T604 GPU.

And yes, the Samsung Exynos 5 Dual is the very same chipset that powers
Samsung's Chromebook and Google’s Nexus 10.

"This will be the first use of an embedded mobile SoC in HPC, which enables
the Mont-Blanc project to explore the challenges and benefits of deeply
integrated energy-efficient processors and GPU accelerators, compared to
traditional homogeneous multicore systems, and heterogeneous CPU + external
GPU architectures," Ramirez explained.

"The Exynos 5 Dual packs the most powerful ARM processors with a programmable
GPU in a low-power mobile device that would normally be in someone’s pocket
and running on a battery. Its performance density, energy efficiency, and low
market price make it an extraordinary building block for prototyping a new
generation of HPC systems."

During the first year of activities, the Mont-Blanc initiative focused on
successfully deploying an HPC system software stack and full-scale scientific
applications on ARM platforms - proving that ARM-based architectures are
feasible alternatives for HPC.

"Now the efforts [are shifting] towards integration of the Exynos platform on
a HPC solution, [along with] software exploitation of the embedded GPU,"
Ramirez added.