Click here to Skip to main content
15,867,975 members

Survey Results

Parallelising your code. Do you do it?   [Edit]

Survey period: 5 Jul 2010 to 12 Jul 2010

Our CPUs aren't getting as fast as fast as they used to, and we're now well and truly in a multi-core world. What do you do to take advantage of this? (Suggested by El Corazon)

OptionVotes% 
I specifically write code that runs in parallel22935.95
I use a language and compiler that supports parallisation automatically ( eg Intel compilers)558.63
I use libraries (eg PLINQ) that provide parallisation support8613.50
I use a framework or runtime (eg Parallel Extensions for .NET) that provide parallisation support15724.65
I let my operating system do what it can33953.22
I let my hardware do what it can18529.04
Not even my hardware can help me.497.69
Respondents were allowed to choose more than one answer; totals may not add up to 100%



 
GeneralRe: Parellism and MFC Pin
Marc Clifton6-Jul-10 7:23
mvaMarc Clifton6-Jul-10 7:23 
GeneralRe: Parellism and MFC Pin
El Corazon6-Jul-10 7:37
El Corazon6-Jul-10 7:37 
GeneralRe: Parellism and MFC Pin
Marc Clifton6-Jul-10 7:48
mvaMarc Clifton6-Jul-10 7:48 
GeneralRe: Parellism and MFC Pin
peterchen9-Jul-10 4:15
peterchen9-Jul-10 4:15 
GeneralBad old threads Pin
Nemanja Trifunovic5-Jul-10 7:15
Nemanja Trifunovic5-Jul-10 7:15 
GeneralOnly Specific Parts of code and only if required. Pin
GPUToaster™4-Jul-10 22:49
GPUToaster™4-Jul-10 22:49 
GeneralRe: Only Specific Parts of code and only if required. Pin
peterchen4-Jul-10 22:58
peterchen4-Jul-10 22:58 
GeneralRe: Only Specific Parts of code and only if required. [modified] Pin
GPUToaster™5-Jul-10 0:31
GPUToaster™5-Jul-10 0:31 
AFAIK: CUDA is strictly for nvidia GPU's only. Earlier versions supported emulators for non-nvidia devices, eliminated completely in v3.0. For using it on others, OpenCL is there (There's both CPU/GPU SIMD implementations). But I prefer CUDA because its faster even than OpenCL and very similar to OpenCL syntax. On top of it NVIDIA/ATI/(to some extent on S3G) has OpenCL support. Moreover nvidia customers as compared to ATI are more. Though last 6 months ATI customer base has shown excellent increase.

To Run CUDA, nvidia 8-Series and above cards are required. I work with GeForce 8500GT (With Compute Capability 1.1) which is very low end card and the performance is little more than my Intel Core 2 Duo E8400 @3.00GHz.

modified on Monday, July 5, 2010 6:59 AM

GeneralRe: Only Specific Parts of code and only if required. Pin
El Corazon6-Jul-10 5:54
El Corazon6-Jul-10 5:54 
Generalparellelism myth PinPopular
ed welch4-Jul-10 22:14
ed welch4-Jul-10 22:14 
GeneralRe: parellelism myth Pin
peterchen4-Jul-10 22:50
peterchen4-Jul-10 22:50 
GeneralRe: parellelism myth Pin
ed welch5-Jul-10 1:05
ed welch5-Jul-10 1:05 
GeneralRe: parellelism myth Pin
Festering5-Jul-10 13:29
Festering5-Jul-10 13:29 
GeneralRe: parellelism myth Pin
El Corazon6-Jul-10 5:15
El Corazon6-Jul-10 5:15 
GeneralI let my operating system do what it can Pin
koolprasad20034-Jul-10 19:41
professionalkoolprasad20034-Jul-10 19:41 

General General    News News    Suggestion Suggestion    Question Question    Bug Bug    Answer Answer    Joke Joke    Praise Praise    Rant Rant    Admin Admin   

Use Ctrl+Left/Right to switch messages, Ctrl+Up/Down to switch threads, Ctrl+Shift+Left/Right to switch pages.