Re: [xmlblaster-devel] Connections/CallbackServer/Subscriptions

[Peter Antman <pra at mint.se>]

Hi,
have to go home now. Will comeback tomorrow. I get verry eratic results.
Sometimes 30 m/s for many connections with RMI and IOR, sometimes 2. And
at ceratain point, the server completely stops sending any messages.

Al meassurments is stolen from TestSubManyClients.

And: the tests are run on the same machine (dual 1.7Ghz), but server
runs in one vm, tests in another. And yes, RPC problem is sockets are
accupied. And for SOCKET 0 is that it was less than 1 msg/sec.

See you.
//Peter


On 26 Sep, Marcel Ruff wrote:
> Peter Antman wrote:
> 
>>Some initial finding on a stress test wich tests
>>a) Many subscribers in one connection.
>>b) Many subscribers with one connection each.
>>
>>Here are the first result with 500 subscriptions ;-)
>>
>>Server memory per subscription consumed
>>             IOR	       SOCKET			RMI			XMLRPC
>>       oneCon  manyCon	       oneCon  manyCon		oneCon  manyCon		oneCon  manyCon
>>100    2631    27484
>>500    1785    26602		1081	18346		1041	22658		452	-
>>
> Strange, the oneCon should be more or less similar for each protocol.
> Did you run gc() three, four times (wiht a sleep(1000L) inbetween) 
> before calculating?
> 
>>
>>Logins/sec
>>	     IOR		SOCKET			RMI			XMLRPC
>>       oneCon  manyCon		oneCon  manyCon		oneCon  manyCon		oneCon  manyCon
>>100    124     2		
>>500    142     2		7	0		167	2		99	-
>>  
>>
> What is measured here??
> 
>>Message/second updated
>>              IOR	       SOCKET			RMI			XMLRPC
>>       oneCon  manyCon	       oneCon  manyCon		oneCon  manyCon		oneCon  manyCon
>>100    35     27
>>500    20     2			11	0		21	2		-	-
>>
> What is the error with XmlRpc? Is the callback port occupied?
> 
> Why has SOCKET 0? Did it fail?
> 
> Are clients running in same JVM or on same computer in different process?
> 
>>
>>
>>Pretty interesting. The XMLRPC did not run at all. And the most
>>effective seems to actully be to have many subscriptions on one Corba
>>connections, with RMI as a competitor. I will do larger test, the just
>>take sooooo long to complete.
>>
> When you have finished your tests are you willing to donate the results
> into performance.html on our homepage?
> 
> thanks
> 
> Marcel
> 
>>
>>//Peter
>>
>>  
>>

-- 
------------------------------------------------------------
Peter Antman	Chief Systems Architect, Business Development
Technology in Media, Box 34105 100 26 Stockholm
WWW: http://www.tim.se	WWW: http://www.backsource.org
Email: pra at tim.se	 
Phone: +46-(0)8-506 381 11 Mobile: 070-675 3942 
------------------------------------------------------------