Use huge parts of RAM

[info]

Hello,
The problem is to acquire image data and process them in
parallel on a Windows 2000 system.

So I am thinking of two threads. First thread is puting
all the image data into a std::queue and a second thread
is reading out the image data and is processing them.

Processing the data might be much slower than acquiring
it, that is why the memory consumption can possibly grow
up to 1.5 GByte worst case.

Will i'm facing problem with that solution?
Can OS provide that if i have physicly 2 GByte RAM.

(sorry for beeing slightly off topic, if s.o. think so pls.
recomment an other newsgroup)

bernd

 

[Victor Bazarov]

The problem is to acquire image data and process them in
parallel on a Windows 2000 system.

So I am thinking of two threads. First thread is puting
all the image data into a std::queue and a second thread
is reading out the image data and is processing them.

Processing the data might be much slower than acquiring
it, that is why the memory consumption can possibly grow
up to 1.5 GByte worst case.

Will i'm facing problem with that solution?
Can OS provide that if i have physicly 2 GByte RAM.

(sorry for beeing slightly off topic, if s.o. think so pls.
recomment an other newsgroup)

comp.os.ms-windows.programmer.win32.

 

[Thomas Matthews]

Hello,
The problem is to acquire image data and process them in
parallel on a Windows 2000 system.

So I am thinking of two threads. First thread is puting
all the image data into a std::queue and a second thread
is reading out the image data and is processing them.

Processing the data might be much slower than acquiring
it, that is why the memory consumption can possibly grow
up to 1.5 GByte worst case.

Will i'm facing problem with that solution?
Can OS provide that if i have physicly 2 GByte RAM.

(sorry for beeing slightly off topic, if s.o. think so pls.
recomment an other newsgroup)

bernd

By the way, threads are not parallel computing. The
processor is still only execution instructions in a
serial sequence. Threads are just tasks with less
overhead than full applications. So when the operating
system switches between threads, it doesn't have to
save and restore as much overhead as for applications.

For true parallel processing, you will need another
processor. For example, you could exercise a DMA
controller to aquire the data and place it into
memory. The processor would manipulate the data.
The key issue is coordinating between the two.

For more information, contact a newsgroup that
discusses your platform. The C++ language does not
have any _standard_ facilities for parallel processing
or multi-threading.

--
Thomas Matthews

C++ newsgroup welcome message:
http://www.slack.net/~shiva/welcome.txt
C++ Faq: http://www.parashift.com/c++-faq-lite
C Faq: http://www.eskimo.com/~scs/c-faq/top.html
alt.comp.lang.learn.c-c++ faq:
http://www.raos.demon.uk/acllc-c++/faq.html
Other sites:
http://www.josuttis.com -- C++ STL Library book
http://www.sgi.com/tech/stl -- Standard Template Library