Question 1: The GPL states that if you redistribute the executable, you
have to redistribute the source. So, if you are using it for your own use,
or internally within one company, you do not have to share the source.
However, if you give it to anyone (and you are not allowed to sell it) you
will have to supply the source as well. NOTE, that since I am not an
employee of UCB, this is merely an informed opinion, not a reflection of
how they see the world.
Question 2: Not that I know of with BOINC. This is really a task for a
different model of computation - the tightly linked multi CPU super
computers that are available (at huge expense), or a moderately linked grid
computation network. You can, however, run tasks on BOINC for which you do
not have the source. There is a wrapper available to wrap executables for
which source is not available. But this would go to a single computer for
each task, not multiple computers.
jm7
Alan Sun
<newsbt...@gmail.
com> To
Sent by: BOINC Dev Mailing List
<boinc_dev-bounce <[email protected]>
[email protected] cc
u>
Subject
[boinc_dev] Two questions, one
10/14/2010 06:06 about the BOINC license, the other
AM is related to the technique.
Hello,
I have 2 questions here, the first one is about the BOINC license. For
example, if I setup a company to provide the BOINC grid computing service,
and I will modify the BOINC source code for customization, on this
condition, should I have to open the source code which I modified based on
the BOINC project? I ask this question because I notice the BOINC project
is
under the GPL license, and according to the GPL, if I did any
modifications,
I have to open the source code of these modifications. Is there anyone can
help me on this?
The second question is related to the computing resources. As we know, the
BOINC system can distribute the tasks to multiple machines, and collect the
calculation result back. This computing mechanism is good, but sometimes, I
got some huge programs, everyone of them has a big calculation work and
needs enormous RAM to run. And, I don't have its source code, so I cannot
replace its original input / output interfaces to the BOINC's. In this
situation, if BOINC only sends it to a single machine (computing node), it
will cost a long time to get the result, or even doesn't work (RAM is not
enough). So, the question is: is there anyway can collect the computing
resource of every single machine to an virtual super computer?
Thank you!
BR
SUN BO TAO
_______________________________________________
boinc_dev mailing list
[email protected]
http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev
To unsubscribe, visit the above URL and
(near bottom of page) enter your email address.
_______________________________________________
boinc_dev mailing list
[email protected]
http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev
To unsubscribe, visit the above URL and
(near bottom of page) enter your email address.
