execution speeds

[Steve Hawkwind]

Hello,

Would any of you say that Java programs executed server-side typically run
faster than those executed client side?

Thanks

 

[Andrew Thompson]

Steve Hawkwind wrote:
...

Would any of you say that Java programs executed server-side typically run
faster than those executed client side?

Sure, why not (shrugs).

Especially if the server is the latest Cray, and
my current desktop PC, remains a 1.8GHz AMD.

Andrew T.

 

[Matt Humphrey]

Hello,

Would any of you say that Java programs executed server-side typically run
faster than those executed client side?

I'll take the other side of the coin from Andrew . The speed of Java
program has much to do with the processor speed, available memory, what else
the CPU is doing, program's networking needs and so forth and nothing to do
with which end of the pipe it runs on. There isn't even a clear meaning for
"server side" as any two machines may both be clients or servers.

Are you trying to optimize some particular architecture? What problem are
you trying to solve?

Matt Humphrey (e-mail address removed) http://www.iviz.com/

 

[Daniel Dyer]

Hello,

Would any of you say that Java programs executed server-side typically
run
faster than those executed client side?

Thanks

Unless you are refering to the client and server Hotspot VM options (i.e.
the -server switch), there is no difference between server-side and
client-side except perhaps the hardware that the program is running on.
There is nothing to stop you running your client programs with the -server
switch.

With Java 5.0 and earlier, using the -server switch with the VM tends to
make long-running applications (such as server processes) perform much
better, at the expense of slower start-up and a larger memory footprint.
Java 6.0 introduces a new register allocator for the client VM that
allegedly significantly improves performance.

Dan.