I've usually found my race conditions, or stupid multi-threading
mistakes with 2 processors just fine. I'm just trying to wrack my
brain to find one that would be solved with 3+ that wouldn't be found
with 2. Hmmm...
Christian.
On 19-Nov-07, at 12:49 AM, Howard Lewis Ship wrote:
It's not deterministic at all and that's the scary part. Another
option would be to create a virtual multi-core computer that exists to
force unlikely but valid race conditions, to ensure that they are
handled properly. But I'm anything but an academic.
On Nov 18, 2007 12:39 PM, Christian Edward Gruber
<[EMAIL PROTECTED]> wrote:
Wouldn't you get proper multiprocessing behaviour from a dual-core
processor, since the choice of processor that is accessing a given
thread at any point isn't deterministic, nor is thread execution
order? Or would you just find the issues faster on 1000 processors.
christian.
On 18-Nov-07, at 1:47 PM, Howard Lewis Ship wrote:
That would be an interesting research project, a JVM that runs
inside
a virtual hyper-multicore system. That is, a simulation of a, say,
1000 core workstation, used to flesh out synchronization problems
that
won't manifest inside single or dual-core CPUs.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
--
Howard M. Lewis Ship
Partner and Senior Architect at Feature50
Creator Apache Tapestry and Apache HiveMind
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
