That still does not tell us what it is you are trying to do. If you are just trying to produce a stream of text, how could that possibly take 18-19 hours? You are wasting everyone's time!

I can't get into many details since it is still classified until the defence day of the PhD.IdeasVacuum wrote:Hi marcoagpinto
That still does not tell us what it is you are trying to do. If you are just trying to produce a stream of text, how could that possibly take 18-19 hours? You are wasting everyone's time!
I am giving credit to PB and to the forum in my thesis.IdeasVacuum wrote:Hi marcoagpinto
Should this forum be coding your Ph.D. for you?
It doesn't fit in RAM, so I divided in multiple stages.Fig wrote:Are you sure every data fit in Ram ? Because the bottleneck may be disk access.
Thank you!Fig wrote:Well if you manage to divide the data in one bench that fit in Ram, then you just have to divide this bench in 4 or 8 smaller parts
Then you launch at the same time, 4 or 8 programs that process each part of data.
That way you will work on 4-8 cores.
Else, you can do the same with one program and 4-8 multithread procedures. But if you are not at ease with read/write, semaphore and so on, try first solution.