• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

German Scientists Tap NVIDIA GPUs To Unlock Secrets Of The Brain


Editor & Senior Moderator
Staff member
Oct 9, 2007
34,496 (9.18/day)
Hyderabad, India
NVIDIA today announced that its GPUs will be used by scientists at Germany's Forschungszentrum Jülich, which hosts the Jülich Supercomputing Centre, one of Europe's largest and most powerful supercomputing resources, to accelerate advanced neurological research targeted at unlocking secrets of the human brain.

NVIDIA also announced a new, multiyear collaboration with the center to drive the next generation of GPU-accelerated scientific research in neuroscience and a range of other fields, including astronomy, astrophysics, material science, particle physics, and protein folding. Together the two organizations are launching the "NVIDIA Application Lab," a jointly run and staffed resource for the European scientific community located at the center's facilities in Jülich.

The lab will enable scientists across Europe to take advantage of GPU-accelerated supercomputing by providing optimized scientific applications and technical support.

"Jülich is one of the most influential supercomputing facilities in Europe, with an impressive track record of solving some of the most challenging scientific problems," said Steve Scott, chief technology officer of the Tesla business at NVIDIA. "The new application lab focused on the breakthrough advantages of GPUs will further enhance their position as one of the world's foremost institutions driving the next wave of scientific discovery."

Advanced Brain Research at Jülich
Neuroscience is among the most exciting and increasingly important research focus areas at Forschungszentrum Jülich. The center is undertaking a new approach to advanced neuroscience research, and potentially uncovering the causes and treatments for autism, multiple sclerosis, Alzheimer's, and other debilitating neurological diseases.

Researchers from the Jülich Institute of Neuroscience and Medicine (INM-1, Structural and Functional Organization of the Human Brain), also at Forschungszentrum Jülich, are using NVIDIA Tesla GPUs to accelerate by as much as 50x the reconstruction of histological brain sections necessary for the rendering of a high-definition, structurally accurate and realistic model of the human brain. Once fully developed, the model will give researchers a previously unattainable level of visibility into brain architecture, function and interconnections with levels of detail never before available to neuroscientists.

To create this model, researchers at the institute INM-1 are reconstructing a vast collection of data sets including images of histological sections (microscopic tissue structure), magnetic resonance images and images from an advanced 3D polarized light imaging (3D-PLI) technique developed at INM-1. 3D-PLI provides, for each voxel of the brain, information about the direction and the inclination of fiber tracts. To trace the tracts over long distances, tractography algorithms are applied, which also require high-performance GPUs.

"3D-PLI is the only way to achieve highly detailed images of nerve fibers in adult human brains, but reconstructing and rendering them in real time into the world's first micro-atlas of the human brain poses a major computational problem," said Professor Katrin Amunts, director of INM-1. "Imagine the billions of nerve cells inside the human brain, connected via fibers. This gives you a sense of the complexity and intricacy needed to accurately model the network within the human brain."

Jülich hopes to leverage key learnings from its neuroscience research to serve as a blueprint for the NVIDIA Application Lab to enable other advanced GPU-accelerated research projects across a range of scientific fields.

Enabling Hundreds of European Scientists - New NVIDIA Application Lab
Beginning later this month, the NVIDIA Application Lab will focus on enabling hundreds of scientists across Europe, including members of the PRACE high performance computing organization, to take advantage of GPU-accelerated systems at the Jülich Supercomputing Centre by providing optimized scientific applications and technical support.

"The new lab will streamline the process of setting up and optimizing new and existing scientific applications to take advantage of GPU acceleration," said Professor Thomas Lippert, director of the Jülich Supercomputing Centre. "This agreement will enable hundreds of researchers to more easily access the game-changing power of GPU computing to advance all types of research."

Researchers in and outside of Jülich are using Jülich's GPU-accelerated supercomputers, including the 206-node Jülich Dedicated GPU Environment (JuDGE) system equipped with NVIDIA Tesla GPUs, which delivers approximately 240 teraflops of peak performance.
Jun 26, 2008
2,303 (0.66/day)
Processor AMD FX-4350
Motherboard Gigabye GA-970A-D3P
Cooling Enermax ETD-T60-VD - 7 x 120mm case fans
Memory Kingston HyperX 2x4GB@1600mhz
Video Card(s) Asus R9 380
Storage Kingston V300 240GB SSD + WD Green 2TB
Display(s) ACER K212HL 27" + Haier 55"
Case Enermax Ostrog (Red)
Power Supply EVGA SuperNOVA B2 750W
Software Win10 64bit
I thought they already did that?

Apr 6, 2009
174 (0.05/day)
Processor 4690K @ 4.2GHz
Motherboard Asus Z97-P
Cooling CRYORIG H7
Memory 16GB (2x8GB) HyperX Fury, DDRS 1600MH2
Video Card(s) EVGA GTX 1070 SC Black Edition
Storage Lots
Display(s) 24" Predator GN246HL, 1080p, 144Hz
Case Phanteks P300
Power Supply Corsair CX650
Software Windows 10 Pro
Benchmark Scores 1/0 0.o
Try a female brain and those über-computers will blow up.
Sep 15, 2007
2,751 (0.73/day)
Police/Nanny State of America
System Name More hardware than I use :|
Processor 4.7 8350 - 4.2 4560K - 4.4 4690K
Motherboard Sabertooth R2.0 - Gigabyte Z87X-UD4H-CF - AsRock Z97M KIller
Cooling Mugen 2 rev B push/pull - Hyper 212+ push/pull - Hyper 212+
Memory 16GB Gskill - 8GB Gskill - 16GB Ballistix 1.35v
Video Card(s) Xfire OCed 7950s - Powercolor 290x - Oced Zotac 980Ti AMP! (also have two 7870s)
Storage Crucial 250GB SSD, Kingston 3K 120GB, Sammy 1TB, various WDs, 13TB (actual capactity) NAS with WDs
Display(s) X-star 27" 1440 - Auria 27" 1440 - BenQ 24" 1080 - Acer 23" 1080
Case Lian Li open bench - Fractal Design ARC - Thermaltake Cube (still have HAF 932 and more ARCs)
Audio Device(s) Titanium HD - Onkyo HT-RC360 Receiver - BIC America custom 5.1 set up (and extra Klipsch sub)
Power Supply Corsair 850W V2 - EVGA 1000 G2 - Seasonic 500 and 600W units (dead 750W needs RMA lol)
Mouse Logitech G5 - Sentey Revolution Pro - Sentey Lumenata Pro - multiple wireless logitechs
Keyboard Logitech G11s - Thermaltake Challenger
Software I wish I could kill myself instead of using windows (OSX can suck it too).
Nvidia is spending big bucks to force their standard into use.

OpenCL or go home.
Dec 22, 2011
2,121 (0.95/day)
System Name Zimmer Frame Rates
Processor Intel i7 920 @ Stock speeds baby
Motherboard EVGA X58 3X SLI
Cooling True 120
Memory Corsair Vengeance 12GB
Video Card(s) Palit GTX 980 Ti Super JetStream
Storage Of course
Display(s) Crossover 27Q 27" 2560x1440
Case Antec 1200
Audio Device(s) Don't be silly
Power Supply XFX 650W Core
Mouse Razer Deathadder Chroma
Keyboard Logitech UltraX
Software Windows 10
Benchmark Scores Epic
Nah they are making big bucks with these deals.

Thanks for the base code Linus Torvalds.


Wile E

Power User
Oct 1, 2006
24,318 (5.89/day)
System Name The ClusterF**k
Processor 980X @ 4Ghz
Motherboard Gigabyte GA-EX58-UD5 BIOS F12
Cooling MCR-320, DDC-1 pump w/Bitspower res top (1/2" fittings), Koolance CPU-360
Memory 3x2GB Mushkin Redlines 1600Mhz 6-8-6-24 1T
Video Card(s) Evga GTX 580
Storage Corsair Neutron GTX 240GB, 2xSeagate 320GB RAID0; 2xSeagate 3TB; 2xSamsung 2TB; Samsung 1.5TB
Display(s) HP LP2475w 24" 1920x1200 IPS
Case Technofront Bench Station
Audio Device(s) Auzentech X-Fi Forte into Onkyo SR606 and Polk TSi200's + RM6750
Power Supply ENERMAX Galaxy EVO EGX1250EWT 1250W
Software Win7 Ultimate N x64, OSX 10.8.4
Nvidia is spending big bucks to force their standard into use.

OpenCL or go home.
OpenCL is behind in development compared to CUDA. Tell khronos to step it up. Until then, CUDA is the better api. They just move too slow.
Sep 15, 2011
4,408 (1.90/day)
Processor Intel Core i7 3770k @ 4.3GHz
Motherboard Asus P8Z77-V LK
Memory 16GB(2x8) DDR3@2133MHz 1.5v Patriot
Video Card(s) MSI GeForce GTX 1080 GAMING X 8G
Storage 59.63GB Samsung SSD 830 + 465.76 GB Samsung SSD 840 EVO + 2TB Hitachi + 300GB Velociraptor HDD
Display(s) Acer Predator X34 3440x1440@100Hz G-Sync
Audio Device(s) Creative X-Fi Titanium PCIe
Power Supply Corsair 850W
Mouse Anker
Software Win 10 Pro - 64bit
Benchmark Scores 30FPS in NFS:Rivals