GPU compiler could "turn desktops into supercomputers"
By Stewart Mitchell
Posted on 18 May 2010 at 10:45
A software team at North Carolina State University claims it could turn a desktop PC into a supercomputer by harnessing the power of its graphics processing unit (GPU).
An average CPU boasts between 20-60 gigaflops of computing power, or 20 to 60 billion operations per second, but that pales compared to some modern GPUs that can run 1 teraflop (a thousand billion operations per second).
The North Carolina researchers have written a software compiler that they say can fully exploit the power of the GPU.
“We have developed a software tool that takes computer program A and translates it into computer program B – which ultimately does the same thing that program A does, but does it more efficiently on a GPU,” said Dr Huiyang Zhou, an associate professor of electrical and computer engineering at North Carolina.
The researchers found that programs designed for conventional processors ran 30% faster using the compiler and a GPU, and that could be the tip of the iceberg. “Tapping into your GPU can turn your personal computer into a supercomputer,” Zhou claimed.
According to the programmers, the main problem for GPU computation is that the architecture of the GPU is designed to process graphics and focuses on turning data into millions of pixels on a screen, with those operations taking place in isolation from each other.
Zhou says his team's compiler effectively translates generic software into a program that runs efficiently on this architecture.
Is your business a social business? For helpful info and tips visit our hub.
Not holding my breath
So the GPU has 20x power and if you recompile your code it runs 30% faster!
GPU's have multiple parallel cores; till we can work out how to do parallel processing, we'll never leverage the true power.
By milliganp on 18 May 2010
The sentence "... programs designed for conventional processors ran 30% faster using the compiler and a GPU ..." seems erroneous.
The report on New Caroline State U website says this instead: "... programs translated by their compiler software ran approximately 30 percent more quickly than those optimized by the GPU developers ... up to 128 times speedups over the naive versions" http://news.ncsu.edu/releases/wmszhoucompiler/
So yes, it is fast.
By mnj_lim on 18 May 2010
- CeBit 2014 diary: Cameron comes to town
- The 5 most interesting UK businesses at SXSW
- Quickest way to upload 1GB? Hop on a train
- Move over Delia: IBM Watson is cooking tonight
- Eric Schmidt on the double-edged smartphone: friend and foe
- Getty joins the race to the bottom
- Hour of Code: five steps to learn how to code
- Sony Xperia Z2 Tablet review: first look
- Sony Xperia Z2 review: first look
- Samsung Galaxy Gear 2 review: first look
- Headings vs headers: how to use both in Word
- Windows Server 2012 R2: how the Datacenter edition could change SMBs
- Invoices and VAT: how to set up your documents correctly
- Nexus 5 vs Samsung Galaxy S4 Active: the best phone for avoiding screen burn
- How much is a social user worth?
- The key to choosing a secure password
- Thunderbolt Bridge: a fast Mac migration tool
- Should you advertise on Twitter?
- How to track a lost smartphone
- Self-publishing success: the best way to sell your book