Re: Unexpected low CPU usage

Johan,

Thanks for starting an interesting thread.

On Wed, 23 Jan 2008 08:48:15 +0100, Johan Stuyts <[email protected]>
wrote:
[snip]
>
>> 2. I'm assuming the following:
>
>>         - you are not using HTTPS
>
>No, I am using HTTPS.

Hopefully a non-issue in 2008, but HTTPS with keepalive is broken on the Sun
1.3.x JVMs

>
>> In your first email you mentioned that your goal is to test the overhead
>> of your framework. That won't be realized with this setup, which isn't
>> really simulating a physical network connection. To understand the
>> overhead the test physical architecture, and specifically the network
>> latency should be roughly comparable. Published studies that compare the
>> performance of Soap and RMI typically report an order of magnitude
>> difference. I don't understand why anyone would choose to use Soap if
>> they have control of both client and server - choosing complexity,
>> bloat, and immature test tools.
>
>I understand I need to run many more tests using different scenarios to
>test all aspects of my server.

That isn't my point. Running a test with the client and server on the same
host does not test the overhead of the framework. Instead of using a network
adapter communication is optimized to use the loopback adapter. If for
example, SOAP were measured to be 20% slower than RMI, we could not   state
anything about relative performance even whether Soap is slower than RMI.

>
>RMI is limited to Java. The advantage of SOAP over XML-RPC is that XFire
>generates a nice WSDL which can be used to generate client code in for
>example .Net.

Thats all true, but only important if you do not have control of the client.
SOAP is a great solution for the EBAY or Amazon's of this world with  of
potential client developers in thousands of organizations.

Unfortunately the "Golden Hammer rule" has led to it also being used as an
enterprise connectivity solution. This benefits hardware vendors much more
than those who use SOAP.


>The real reason is I just don't want to force somebody to use a particular
>protocol to be able to use my application. I want the application to be as
>accessible to developers/administrators/hobbyists/... as possible.

Do you expect users to write their own client code or will you make a
library available?

>
>As I sent in another post I found the configuration option that allows me
>to disable the delay. I have already run successful tests with my SOAP
>client. It is now doing 800+ roundtrips per second instead of 60+.

Let us know what the numbers are when we're communicating across a network
with a realistic latency. It would be interesting to see how the numbers
compare for different protocols.

===================================
This list is hosted by DevelopMentor®  http://www.develop.com

View archives and manage your subscription(s) at http://discuss.develop.com