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]

Reply via email to