Overclock.net › Forums › Industry News › Software News › [PCPer] NVIDIA Updates CUDA: Major Release for Science Research...
New Posts  All Forums:Forum Nav:

[PCPer] NVIDIA Updates CUDA: Major Release for Science Research...

post #1 of 11
Thread Starter 
Quote:
NVIDIA has traditionally been very interested in acquiring room in the high-performance computing for scientific research market. For a lot of functions, having a fast and highly parallel processor saves time and money compared to having a traditional computer crunch away or having to book time with one of the world’s relatively few supercomputers. Despite the raw performance of a GPU, adequate development tools are required to bring the simulation or calculation into a functional program to execute on said GPU. NVIDIA is said to have had a strong lead with their CUDA platform for quite some time; that lead will likely continue with releases the size of this one.

Source

This is pretty interesting... I am not really familiar with nVidia CUDA...could somone maybe elaborate on it a little.
FUSION
(13 items)
 
  
CPUMotherboardRAMHard Drive
AMD Llano A8-3870k ASUS F1A75-M Pro FM1 uATX Corsair Vengeance Corsair Force 3 
Hard DriveOptical DriveCoolingOS
Hitachi ASUS DVD Burner Cooler Master Hyper TX3 Windows 7 Ultimate 64-bit 
MonitorKeyboardPowerCase
ASUS VE247H 24" 1080p Razer Lycosa Corsair CX500 Fractal Define Design R3 ATX 
Mouse
Razer DeathAdder 
  hide details  
Reply
FUSION
(13 items)
 
  
CPUMotherboardRAMHard Drive
AMD Llano A8-3870k ASUS F1A75-M Pro FM1 uATX Corsair Vengeance Corsair Force 3 
Hard DriveOptical DriveCoolingOS
Hitachi ASUS DVD Burner Cooler Master Hyper TX3 Windows 7 Ultimate 64-bit 
MonitorKeyboardPowerCase
ASUS VE247H 24" 1080p Razer Lycosa Corsair CX500 Fractal Define Design R3 ATX 
Mouse
Razer DeathAdder 
  hide details  
Reply
post #2 of 11
When it says this:

"A new compiler which is based on the LLVM infrastructure, making good on their promise to open the CUDA platform to other architectures -- both software and hardware"


I'm pretty sure that they don't mean that it's being opened to running on AMD/ATI graphics cards, right? But what other architectures would it run on? Surely not x86 or ARM? I think I understand the software bit - LLVM being (as far as I can work out) a different branch of compilers. If it means it will run on other GPUs, then this is huge news. Otherwise, it just seems like an incremental update?
Little Messy
(16 items)
 
The 1720 (Laptop)
(13 items)
 
 
CPUMotherboardGraphicsRAM
Phenom II x4 955 (@ 3.8) MSI 890GXM-G65 ASUS HD 6850 (@ 975/1125) G.Skill RipjawX 
Hard DriveHard DriveOptical DriveCooling
Samsung Eco Green Samsung F3 Sony DVD+-RW Writer CoolerMaster Hyper 212+ 
OSOSMonitorKeyboard
Windows 7 Home Premium Arch Linux Dell ST2420 Microsoft Reclusa 
PowerCaseMouse
OCZ StealthXStream II 600W NZXT Vulcan Logitech MX518 
CPUMotherboardGraphicsRAM
Core 2 Duo T8300 Dell's own 8600M 4.0GB Hyundai Electronics DDR2 
Hard DriveOS
250GB Hitachi / 250GB Samsung Windows Vista 32 bit 
  hide details  
Reply
Little Messy
(16 items)
 
The 1720 (Laptop)
(13 items)
 
 
CPUMotherboardGraphicsRAM
Phenom II x4 955 (@ 3.8) MSI 890GXM-G65 ASUS HD 6850 (@ 975/1125) G.Skill RipjawX 
Hard DriveHard DriveOptical DriveCooling
Samsung Eco Green Samsung F3 Sony DVD+-RW Writer CoolerMaster Hyper 212+ 
OSOSMonitorKeyboard
Windows 7 Home Premium Arch Linux Dell ST2420 Microsoft Reclusa 
PowerCaseMouse
OCZ StealthXStream II 600W NZXT Vulcan Logitech MX518 
CPUMotherboardGraphicsRAM
Core 2 Duo T8300 Dell's own 8600M 4.0GB Hyundai Electronics DDR2 
Hard DriveOS
250GB Hitachi / 250GB Samsung Windows Vista 32 bit 
  hide details  
Reply
post #3 of 11

Just when I thought CUDA was beginning to slip they rocket-propel themselves forward! I can't wait to try out the LLVM compiler!

post #4 of 11
Quote:
Originally Posted by Korlus View Post

When it says this:
"A new compiler which is based on the LLVM infrastructure, making good on their promise to open the CUDA platform to other architectures -- both software and hardware"
I'm pretty sure that they don't mean that it's being opened to running on AMD/ATI graphics cards, right? But what other architectures would it run on? Surely not x86 or ARM? I think I understand the software bit - LLVM being (as far as I can work out) a different branch of compilers. If it means it will run on other GPUs, then this is huge news. Otherwise, it just seems like an incremental update?

NVIDIA is releasing (or has released?) the source for the compiler. This will allow developers to add new architecture or language support that is not currently present. X86 is already supported but I suppose new architectures could mean ARM of AMD GPUs.
 
HTPC
(8 items)
 
 
CPUMotherboardGraphicsRAM
i7-3930k MSI X79MA-GD45 Powercolor 7950 G Skill 
Hard DriveCoolingCoolingCooling
Samsung 830 Swiftech Apogee XT Rev. 1 EK-FC HWLabs Black Ice GTX 240 and a Koolance HX-CU72... 
CoolingCoolingOSOS
MCP 35X and a MCP 350 Koolance 401x2 Opensuse 12.1 Windows 7 
KeyboardPowerCase
Ducky OCN with cherry blues Antec True Power New 650W Cooler Master CM690 II 
CPUMotherboardGraphicsRAM
i5-750 Gigabyte p55-ud2 GTX 430 G Skill 
Hard DriveOSPowerCase
Samsung HD204UI Fedora 15 Corsair CX-430 Silverstone GC04 
  hide details  
Reply
 
HTPC
(8 items)
 
 
CPUMotherboardGraphicsRAM
i7-3930k MSI X79MA-GD45 Powercolor 7950 G Skill 
Hard DriveCoolingCoolingCooling
Samsung 830 Swiftech Apogee XT Rev. 1 EK-FC HWLabs Black Ice GTX 240 and a Koolance HX-CU72... 
CoolingCoolingOSOS
MCP 35X and a MCP 350 Koolance 401x2 Opensuse 12.1 Windows 7 
KeyboardPowerCase
Ducky OCN with cherry blues Antec True Power New 650W Cooler Master CM690 II 
CPUMotherboardGraphicsRAM
i5-750 Gigabyte p55-ud2 GTX 430 G Skill 
Hard DriveOSPowerCase
Samsung HD204UI Fedora 15 Corsair CX-430 Silverstone GC04 
  hide details  
Reply
post #5 of 11
Quote:
Originally Posted by mdatmo View Post


NVIDIA is releasing (or has released?) the source for the compiler. This will allow developers to add new architecture or language support that is not currently present. X86 is already supported but I suppose new architectures could mean ARM of AMD GPUs.


The LLVM compiler is also providing a 10% performance boost to existing programs. Something for everyone. I'm personally fine with using C++.

post #6 of 11
Quote:
Originally Posted by mikezachlowe2004 View Post

Quote:
NVIDIA has traditionally been very interested in acquiring room in the high-performance computing for scientific research market. For a lot of functions, having a fast and highly parallel processor saves time and money compared to having a traditional computer crunch away or having to book time with one of the world’s relatively few supercomputers. Despite the raw performance of a GPU, adequate development tools are required to bring the simulation or calculation into a functional program to execute on said GPU. NVIDIA is said to have had a strong lead with their CUDA platform for quite some time; that lead will likely continue with releases the size of this one.
Source
This is pretty interesting... I am not really familiar with nVidia CUDA...could somone maybe elaborate on it a little.

as i understand cuda is basically a software platform for writing code to run directly on nvidia gpus. highly-parallel tasks run much much faster thanks to the GPU architecture. things like video encoding/editing, physics processing in games, and folding all run better with cuda.


actually what nvidia is doing is it's moving to a gpgpu system - one gpu-based chip for the entire computer, no need for a CPU. so we have intel releasing CPU's with integrated GPU's(which accelerate video encoding), AMD releasing CPU's with very fast integrated GPU's as well, and nvidia releasing a GPU capable of performing x86 tasks. Will be interesting to see how everything develops in the coming years
post #7 of 11
I don't see much point in this. Except maby putting nVidia into worse situation with the GPU computing.
Would be much clever to move to and support existing cross computing platform OpenCL.
post #8 of 11
whoa justwhen intel thought it was king, little ol nvidia might mess things up in years to come ey?
GOKU
(18 items)
 
Vegeta
(14 items)
 
 
CPUMotherboardGraphicsRAM
i7 3770K, @ 4.5 delidded liquid ultra GA-Z77_MX-D3H GTX 1070FE 8Gb Corsair Vengance LP 
Hard DriveHard DriveHard DriveCooling
intel 330 ssd  WD Caviar black 1Tb WD 2.5 500Gb scorpio black XSPC Raystorm CPU  
CoolingCoolingOSMonitor
240mm xspc rad D5 Vario PMP-450 windows 10 PRO Dell u2713hm 
KeyboardPowerCaseMouse
CM Quickfire Storm Seasonic G series 650W Parvum S2.0 CM sentinel II 
Mouse PadAudio
Strike x desk pad Seinhiser HD 435 
CPUMotherboardGraphicsRAM
i7 2600 Asrock H61m-itx Galaxy 750ti slim OC kingston 2x4gb low profile 
Hard DriveHard DriveCoolingOS
SP 256Gb SSD  western digital black caviar 2Tb Noctua NH-L9i Windows 7 PRO x64 
MonitorKeyboardPowerCase
samsung FHD 48 inch logitech wireless with trackpad seasonic G series 650W gold plus UMX-1 plus 
MouseAudio
xbox 360 wireless pioneer reciever + 5.1 surround 
  hide details  
Reply
GOKU
(18 items)
 
Vegeta
(14 items)
 
 
CPUMotherboardGraphicsRAM
i7 3770K, @ 4.5 delidded liquid ultra GA-Z77_MX-D3H GTX 1070FE 8Gb Corsair Vengance LP 
Hard DriveHard DriveHard DriveCooling
intel 330 ssd  WD Caviar black 1Tb WD 2.5 500Gb scorpio black XSPC Raystorm CPU  
CoolingCoolingOSMonitor
240mm xspc rad D5 Vario PMP-450 windows 10 PRO Dell u2713hm 
KeyboardPowerCaseMouse
CM Quickfire Storm Seasonic G series 650W Parvum S2.0 CM sentinel II 
Mouse PadAudio
Strike x desk pad Seinhiser HD 435 
CPUMotherboardGraphicsRAM
i7 2600 Asrock H61m-itx Galaxy 750ti slim OC kingston 2x4gb low profile 
Hard DriveHard DriveCoolingOS
SP 256Gb SSD  western digital black caviar 2Tb Noctua NH-L9i Windows 7 PRO x64 
MonitorKeyboardPowerCase
samsung FHD 48 inch logitech wireless with trackpad seasonic G series 650W gold plus UMX-1 plus 
MouseAudio
xbox 360 wireless pioneer reciever + 5.1 surround 
  hide details  
Reply
post #9 of 11
Quote:
Originally Posted by Pikivirta View Post

I don't see much point in this. Except maby putting nVidia into worse situation with the GPU computing.
Would be much clever to move to and support existing cross computing platform OpenCL.

I'm not sure that's entirely true. CUDA is (in most scenarios) faster than OpenCL for things. While I would like the move to an open language, if CUDA is still faster after being properly implemented on other architectures, it will be nice to see a wider support for AMD in GPGPU scenarios.

If that is the case, then Nvidia will still be adding to CUDA and will still have control of it, but will see wider support for it as other architectures adopt it. I'm not entirely sure how they'll use this, beyond possibly tailoring their architecture to better support it, or something?
Little Messy
(16 items)
 
The 1720 (Laptop)
(13 items)
 
 
CPUMotherboardGraphicsRAM
Phenom II x4 955 (@ 3.8) MSI 890GXM-G65 ASUS HD 6850 (@ 975/1125) G.Skill RipjawX 
Hard DriveHard DriveOptical DriveCooling
Samsung Eco Green Samsung F3 Sony DVD+-RW Writer CoolerMaster Hyper 212+ 
OSOSMonitorKeyboard
Windows 7 Home Premium Arch Linux Dell ST2420 Microsoft Reclusa 
PowerCaseMouse
OCZ StealthXStream II 600W NZXT Vulcan Logitech MX518 
CPUMotherboardGraphicsRAM
Core 2 Duo T8300 Dell's own 8600M 4.0GB Hyundai Electronics DDR2 
Hard DriveOS
250GB Hitachi / 250GB Samsung Windows Vista 32 bit 
  hide details  
Reply
Little Messy
(16 items)
 
The 1720 (Laptop)
(13 items)
 
 
CPUMotherboardGraphicsRAM
Phenom II x4 955 (@ 3.8) MSI 890GXM-G65 ASUS HD 6850 (@ 975/1125) G.Skill RipjawX 
Hard DriveHard DriveOptical DriveCooling
Samsung Eco Green Samsung F3 Sony DVD+-RW Writer CoolerMaster Hyper 212+ 
OSOSMonitorKeyboard
Windows 7 Home Premium Arch Linux Dell ST2420 Microsoft Reclusa 
PowerCaseMouse
OCZ StealthXStream II 600W NZXT Vulcan Logitech MX518 
CPUMotherboardGraphicsRAM
Core 2 Duo T8300 Dell's own 8600M 4.0GB Hyundai Electronics DDR2 
Hard DriveOS
250GB Hitachi / 250GB Samsung Windows Vista 32 bit 
  hide details  
Reply
post #10 of 11
Quote:
Originally Posted by Korlus View Post

When it says this:
"A new compiler which is based on the LLVM infrastructure, making good on their promise to open the CUDA platform to other architectures -- both software and hardware"
I'm pretty sure that they don't mean that it's being opened to running on AMD/ATI graphics cards, right? But what other architectures would it run on? Surely not x86 or ARM? I think I understand the software bit - LLVM being (as far as I can work out) a different branch of compilers. If it means it will run on other GPUs, then this is huge news. Otherwise, it just seems like an incremental update?

CUDA is open for other architectures, so long as those pay Nvidia a licensing fee, and AMD isn't about to pay for a closed standard developed by their competition when they can use an open-standard for free.
Foldatron
(17 items)
 
Mat
(10 items)
 
Work iMac
(9 items)
 
CPUMotherboardGraphicsGraphics
i7 950 EVGA x58 3-way SLI EVGA GTX 660ti GTX 275 
RAMHard DriveHard DriveHard Drive
3x2GB Corsair Dominator DDR3-1600 80GB Intel X25-M SSD 2TB WD Black 150GB WD Raptor 
Hard DriveOSMonitorKeyboard
2x 150GB WD V-raptor in RAID0 Win7 Home 64-bit OEM 55" LED 120hz 1080p Vizio MS Natural Ergonomic Keyboard 4000 
PowerCase
750W PC P&C Silencer CoolerMaster 690 
CPUGraphicsRAMHard Drive
Intel Core i5 2500S AMD 6770M 8GB (2x4GB) at 1333Mhz 1TB, 7200 rpm 
Optical DriveOSMonitorKeyboard
LG 8X Dual-Layer "SuperDrive" OS X Lion 27" iMac screen Mac wireless keyboard 
Mouse
Mac wireless mouse 
CPUGraphicsRAMHard Drive
i7-2600K AMD 6970M 1GB 16GB PC3-10600 DDR3 1TB 7200rpm 
Hard DriveOptical DriveOSMonitor
256GB SSD 8x DL "SuperDrive" OS X 10.7 Lion 27" 2560x1440 iMac display 
Monitor
27" Apple thunderbolt display 
  hide details  
Reply
Foldatron
(17 items)
 
Mat
(10 items)
 
Work iMac
(9 items)
 
CPUMotherboardGraphicsGraphics
i7 950 EVGA x58 3-way SLI EVGA GTX 660ti GTX 275 
RAMHard DriveHard DriveHard Drive
3x2GB Corsair Dominator DDR3-1600 80GB Intel X25-M SSD 2TB WD Black 150GB WD Raptor 
Hard DriveOSMonitorKeyboard
2x 150GB WD V-raptor in RAID0 Win7 Home 64-bit OEM 55" LED 120hz 1080p Vizio MS Natural Ergonomic Keyboard 4000 
PowerCase
750W PC P&C Silencer CoolerMaster 690 
CPUGraphicsRAMHard Drive
Intel Core i5 2500S AMD 6770M 8GB (2x4GB) at 1333Mhz 1TB, 7200 rpm 
Optical DriveOSMonitorKeyboard
LG 8X Dual-Layer "SuperDrive" OS X Lion 27" iMac screen Mac wireless keyboard 
Mouse
Mac wireless mouse 
CPUGraphicsRAMHard Drive
i7-2600K AMD 6970M 1GB 16GB PC3-10600 DDR3 1TB 7200rpm 
Hard DriveOptical DriveOSMonitor
256GB SSD 8x DL "SuperDrive" OS X 10.7 Lion 27" 2560x1440 iMac display 
Monitor
27" Apple thunderbolt display 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Software News
Overclock.net › Forums › Industry News › Software News › [PCPer] NVIDIA Updates CUDA: Major Release for Science Research...