- Als neu kennzeichnen
- Lesezeichen
- Abonnieren
- Stummschalten
- RSS-Feed abonnieren
- Kennzeichnen
- Anstößigen Inhalt melden
Hello,
I would like to use TBB to get better performances on memory allocation and parallelization of disk access on a code that already exists.
I was therefore trying to see what kind of modifications the use of TBB would imply for my code.
I began bygiving alook at the use of parallel_for
parallel_for(blocked_range
it seems that ApplyFoo has to be a class ?
Does this mean that I cannot use the TBB to improve the loops I already have on the codethat has beendevelopped in C ?
Thank you
Link kopiert
2 Antworten
- Als neu kennzeichnen
- Lesezeichen
- Abonnieren
- Stummschalten
- RSS-Feed abonnieren
- Kennzeichnen
- Anstößigen Inhalt melden
There is a separate forum for TBB. Yes, the TBB classes require C++ compilation. You could localize the use of C++ in your application by invoking TBB in C++ functions defined with "extern C." If your application is suitable for OpenMP, as appears to be implied in your consideration of parallel_for, that may be a better method than TBB.
If you need parallel disk access for performance, striped data sets (RAID0) may be a good approach. Intel has announced a solid state disk program (no products yet).
If you need parallel disk access for performance, striped data sets (RAID0) may be a good approach. Intel has announced a solid state disk program (no products yet).
- Als neu kennzeichnen
- Lesezeichen
- Abonnieren
- Stummschalten
- RSS-Feed abonnieren
- Kennzeichnen
- Anstößigen Inhalt melden
Ups, I don't know howI gotredirected on the wrong part of the forum after subscription. I thought I was on the TBB part.
Sorry for that
and thank you for the answer.
Antworten
Themen-Optionen
- RSS-Feed abonnieren
- Thema als neu kennzeichnen
- Thema als gelesen kennzeichnen
- Diesen Thema für aktuellen Benutzer floaten
- Lesezeichen
- Abonnieren
- Drucker-Anzeigeseite