Hi Jiri,
Thanks for the link it clears up better what you and Thomas are saying. You
both have been extreemly helpful. What you and Thomas need to understand is the
typical developer will not ask the questions that I do. To me a profiler and
static code analysis tools are as important as a IDE for productivity and
assuring quality.
I did run visualvm on two other windows PC's (Windows Xp Sp2 and Windows 2000
prof sp4) and did not get the error that I got for the instrumentation profiler
(the sampler did not generate any errors).
Thanks for your time I am trying to get different functional groups interested
in using visualvm as we upgrade to jdk 1.6.0_7 and above so they get the
visualvm profiler.
I am just trying to do what noone else has done and get a system response time
via lan under 1 millisecond with inexpensive hardware and open source software
(Tomcat/JBoss) and need a way to find bottlenecks using a tool that can measure
down to millisecond or below. I am/was at 3-4 milliseconds measure at the
client using JAX-WS and just last night figured out where my remaining
bottleneck (using visualvm and help from the Tomcat group) might be to achieve
my goals.
Just imagine the size of a data center if vendors/developers could create
solutions that responded under a microsecond (I have code that used to be under
1 nanosecond but now around 5 microsends with 3 tier caching). Depending upon
the network technology you could serve up whole countries with that type of
capability using open source like Tomcat no less! Just need the 200Gbps
networks the universities have here in the US :-)
Best Regards,
-Tony
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org
