Incidentally, I'm working on code that allows for toggling of
concurrent map providers in different contexts. We could probably
adapt this to be more general-purpose. However, currently, this is not
something that can happen via our current configuration mechanisms,
since some of the key maps in q
I think this thread shows that there are some critical maps at the
core of JPA, and it shows that the optimal map for an environment is
debatable. Is this something we could make configurable? I'm
personally very very skeptical of micro-benchmarks, and won't be
convinced which is best unt
Two
Patrick Linskey wrote:
> Hi,
>
> How many cores / CPUs were you using in these tests?
>
> -Patrick
>
> On 6/1/07, David Ezzio (asmtp) <[EMAIL PROTECTED]> wrote:
>> Hi Marc,
>>
>> I still was not able to get HighScaleLib to work. It produces a
>> SecurityException when attempting to get the
Hi,
How many cores / CPUs were you using in these tests?
-Patrick
On 6/1/07, David Ezzio (asmtp) <[EMAIL PROTECTED]> wrote:
Hi Marc,
I still was not able to get HighScaleLib to work. It produces a
SecurityException when attempting to get the Unsafe object. I decided to
avoid changing the rele
Hi Marc,
I still was not able to get HighScaleLib to work. It produces a
SecurityException when attempting to get the Unsafe object. I decided to
avoid changing the relevant security policy.
On the other hand, I did test Emory University's backport of the
java.util.concurrent package. This p
