Wednesday, September 9th 2020
GIGABYTE, Northern Data AG and AMD Join Forces to Drive HPC Mega-Project
GIGABYTE Technology, an industry leader in high-performance servers and workstations, today is announcing a partnership with Northern Data AG to create a HPC mega-project with computing power of around 3.1 exaflops. GIGABYTE will supply GPU-based server systems equipped with proven AMD EPYC processors and AMD Radeon Instinct accelerators from technology partner AMD, a leading provider of high performance computing and graphics technologies, to Northern Data.
Northern Data develops a distributed computing cluster based on the hardware at locations in Norway, Sweden and Canada, which in its final stage of deployment will provide FP32 computing power of around 3.1 exaflops (3.1 million teraflops and 274.54 petaflops FP64). The world's fastest supercomputer, the Japanese "Fukagu" (Fujitsu), has a calculation power of 1.07 exaflops FP32 and 415.3 petaflops FP64, whereas the second fastest, the US supercomputer "Summit" (IBM) has a calculation power of 0.414 exaflops FP32 and 148.0 petaflops FP64.
Thomas Yen, Sales and BDM at GIGABYTE Technology, comments: "We are very excited about this mega-project and the lively partnership with Northern Data. In cooperation with AMD, we are able to design, manufacture and deliver high volumes of servers in very high production quality within a few months. This has convinced Northern Data to equip its GPU-based HPC platform, which does rendering, deep learning or artificial intelligence, with GIGABYTE servers."
Sukh Dhillon, EMEA Commercial Channel Sales Director at AMD, comments: "We are delighted to be in collaboration with GIGABYTE and Northern Data, one of the global leaders in HPC computing. We look forward to working on this unique flagship project utilizing second generation AMD EPYC processors and AMD Radeon Instinct accelerators helping set superior standards for performance, security and scalability for the most demanding workloads."
The hardware will be delivered in the coming weeks and months. GIGABYTE's high-performance servers combine AMD processors selected by Northern Data with a dedicated server architecture. In this way, GIGABYTE adapts the CPU- and GPU clusters developed by Northern Data to Northern Data's technical requirements.
Northern Data develops a distributed computing cluster based on the hardware at locations in Norway, Sweden and Canada, which in its final stage of deployment will provide FP32 computing power of around 3.1 exaflops (3.1 million teraflops and 274.54 petaflops FP64). The world's fastest supercomputer, the Japanese "Fukagu" (Fujitsu), has a calculation power of 1.07 exaflops FP32 and 415.3 petaflops FP64, whereas the second fastest, the US supercomputer "Summit" (IBM) has a calculation power of 0.414 exaflops FP32 and 148.0 petaflops FP64.
Sukh Dhillon, EMEA Commercial Channel Sales Director at AMD, comments: "We are delighted to be in collaboration with GIGABYTE and Northern Data, one of the global leaders in HPC computing. We look forward to working on this unique flagship project utilizing second generation AMD EPYC processors and AMD Radeon Instinct accelerators helping set superior standards for performance, security and scalability for the most demanding workloads."
The hardware will be delivered in the coming weeks and months. GIGABYTE's high-performance servers combine AMD processors selected by Northern Data with a dedicated server architecture. In this way, GIGABYTE adapts the CPU- and GPU clusters developed by Northern Data to Northern Data's technical requirements.
2 Comments on GIGABYTE, Northern Data AG and AMD Join Forces to Drive HPC Mega-Project
This makes sense though, especially for AMD as currently, no other entity can offer the same level of integration and act as essentially a "one-stop shop" in providing CPUs and GPUs for the HPC market, as well as their own NVLink style mesh interconnect technology to support it (though from what I've been reading AMD's interconnect is something different than NVLink since it's obviously more heterogeneous) . Nvidia obviously doesn't offer x86 and Intel doesn't offer GPU accelerators currently, although they're trying to change this (though I think it's going to take them several years to reach parity with AMD in this department, especially considering the fact that AMD's revenue is consistently increasing and thus their R&D budget for all markets they compete it, meaning that AMD will only offer increasing tough competition).
Either way, it's pretty neat that a supercomputer arms race is developing that's evolving at an ever increasing rate, awesome.