[Gambas-user] Gambas performances

[Olivier Cruilles]

I measure the time to execute 730 snmpget commands started as process  
(SHELL command) in a Gambas program.


In the first time I start 20 processes in the same time and when one  
is finished, I start a new one to follow the list of requests
to execute.

I measure the time from the first process started to the last process  
finished.

On the Celeron cpu, during all the time, the cpu is at 100%

On the Dual-Core machine, during all the time, the both cpus are at  
around 80%

On the Xeon machine, during all the time, the both cpus are at around  
85%

I try to change the network connection from Ethernet card 100 Mb to  
1Gb but nothing change in all cases.


So it's really about cpu technology where the problem is. Maybe the  
Xeon works really on SMP mode with a x86_64 distribution and
lesser with x86_32 kernel ?

I don't understand why ?

I use the official packages of Gambas on the Fedora 10 distribution.  
Does you optimize the runtime of Gambas for x86_32 dual core cpu ?


Olivier Cruilles
Mail: linuxos at ...1896...


Le 17 août 09 à 21:22, Benoît Minisini a écrit :

>> Hello Benoit,
>>
>> Just a question about performances of Gambas 2.14 on Fedora 10
>> x86_32  (32bits)
>>
>> I'm trying my scheduler written in Gambas on Fedora 10 on 3
>> differentes machines:
>>
>> 1)  1x Celeron 2.6 Ghz    512Mb Ram  (simple PC)
>>
>> 	Time to execute 730 snmp requests:  around 49 sec
>>
>> 2) 1x Dual-Core 2,4 Ghz    512Mb Ram     (simple PC)     (2 cpu  
>> inside
>> the Dual-core)      3500 bogomips per cpu
>>
>> 	Time to execute 730 snmp requests:  around  21 sec
>>
>> 3) 1x Intel Xeon 3,4 GHz   1Gb Ram    (HP DL360 server)     (2 cpu
>> inside the Intel Core)     6800 bogomips per cpu
>>
>> 	Time to execute 730 snmp requests:  around 37 sec
>>
>>
>> I don't understand why this is not better with the Xeon machine. May
>> be I should try with the x86_64 version of the Fedora 64
>> to have a better performances on this machine ?
>>
>> I think the Xeon does not use really the both cpu than then Dual-core
>> in this test ?
>>
>>
>> If you have any idea about that, I will be very happy, because I need
>> to win time to execute
>> much more snmp request and others programs written in Gambas on the
>> same machine.
>>
>> For the moment I have 1 machine per program and I would put all of
>> this programs on a only one.
>>
>>
>>
>> Olivier Cruilles
>> Mail: linuxos at ...1896...
>>
>
> 1) What did you measure exactly?
>
> 2) If a faster CPU has no effect, or a reverse effect, then that  
> means that
> the bottleneck is not in the CPU speed. What is the CPU load?
>
> 3) What is done exactly when you do one snmp request? Do you send  
> and receive
> packets from the network?
>
> Well, I can't say better without more information...
>
> -- 
> Benoît
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008  
> 30-Day
> trial. Simplify your report design, integration and deployment - and  
> focus on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> _______________________________________________
> Gambas-user mailing list
> Gambas-user at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/gambas-user