[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [amibroker] Quad-core test results



PureBytes Links

Trading Reference Links

Hi TJ - it may be trivial but that is what I am best at!   8 - )
 
But seriously, I need to run these on individual tickers because I want to see individual results for all param values rather than "basket results".  I understand that you need to account for what *everyone* is doing, but I just need to account for what I do, and for my work I am very happy with the results. TJ, I only use EOD database, so I think my longest data history is probably about 5,000 bars or so, but if you would like to know results for any other scenarios and would care to send me any sort of "stress test" code or the like, I will be happy to run it with up to 4 instance and report back with the results - just let me know!
 
Steve
----- Original Message -----
Sent: Friday, May 23, 2008 2:37 AM
Subject: Re: [amibroker] Quad-core test results

"For background, the test was an optimization run on a single ticker, which is how I will be running all my other tests ( ticker was QID, 468 EOD data bars ). All tests used the same code with different param settings, all tests had about 46,000 opt steps."
 
These tests are irrelevant. Single ticker END OF DAY 468 bars is ONLY 14976 BYTES of memory. That's 14K or 0.014MB. Fits ENTIRELY in on-chip CPU cache.
On ANY CPU (even the most low-end cpu).
 
The entire discussion was that *SYSTEM* (non-CPU) MEMORY SPEED is the limit
===============================================
and this applies of course to cases when data did NOT FIT in on-chip memory.
 
Single-ticker 468 bar optimization is TOO TRIVIAL to be relevant.
 
Run something serious that requires at least 20MB of RAM so it can't sit in the cache!

Best regards,
Tomasz Janeczko
amibroker.com
----- Original Message -----
Sent: Friday, May 23, 2008 6:25 AM
Subject: RE: [amibroker] Quad-core test results

Steve,

 

Can you provide some information about the differences between the machine you refer to as you faster primary machine and the new quad core ?

 

-          Number of CPU?s

-          Number of Cores

-          On Chip Cache Memory Size

-          Total Memory

-          OS System & Version & SP

 

It would appear that without splitting up the task that you are basically getting a 2.5 times improvement between your faster primary machine and the quad core i.e. 75 minutes .vs. 30 minutes.

 

This could be accounted for by several things in the configuration.

 

I find it more interesting however that you can drive the same task through all four cores on your quad at the same speed as running a single optimization.

 

I would think the implication is that there would be benefit to being able to have AB do this automatically, at least on newer hardware and of course as time goes along more and more hardware out there in the hands of users will be of this caliber.

 

Mark K,

 

The problem with trying to get IO to do something like this is that my understanding of OLE / Automation is that it is only able to see what is the first instance of AB running.  The implication is that while one could be running IO in each of the four cores they would by default all be talking to the same instance of Broker.exe.

 


From: amibroker@xxxxxxxxxxxxxxx [mailto:amibroker@xxxxxxxxxxxxxxx] On Behalf Of Steve Dugas
Sent: Thursday, May 22, 2008 8:50 PM
To: Yahoo - AmiBroker
Subject: [amibroker] Quad-core test results

 

Hi All,

 

I finished setting up the new quad-core machine and ran my first tests today, so am posting the promised test results. All I can say is Wow!  Really nice improvements, even better than I expected to see in my highest hopes. I am happy as a pig in sh*t !!     8 - )

 

For background, the test was an optimization run on a single ticker, which is how I will be running all my other tests ( ticker was QID, 468 EOD data bars ). All tests used the same code with different param settings, all tests had about 46,000 opt steps.

 

To put things in perspective and show why I am so happy, the original code was about 2,400 lines. The first time I ran it, on my old backup computer which I was using as a dedicated optimization machine, it took 7 1/2 hours to run. So then I copied the code and created a shorter version, removing everything which wasn't absolutley necessary for the optimizer, and that reduced the run time to 2 1/2 hours. Then I ran this short version on my faster primary machine and that took 1 1/4 hours, which is about what I was expecting to see on the new machine.

 

So today I started by running just ran one instance on the quad-core - that took only 30 mins and Task Manager showed it was using just 25% of the total processing power!  Well to make a long story short, I kept adding one instance at a time, all instances ran in 30 mins and each used up an additional 25% of the CPU power. In the end, I was easily running 4 simultaneous instances. This pretty much kept the CPU tached at 100% but all instances ran fine, all finished in 30 mins and I didn't experience any problems at all. I was even saving the first ones to spreadsheets while the final ones were still finishing, wow everything just worked flawlessly!  So, the quad-core can run 4 seperate opts simultaneously in 30 mins, which averages out to 7 1/2 minutes per opt, which =

 

10X improvement over running the short code on the fast machine...

20X improvement over running the short code on the old dedicated optimization machine...

60X improvement over running the original code on the old optimization machine...  Awesome !!!

 

To those who were wondering what is the best machine to get for running AB, it looks to me like quad is the way to go.  ( My machine has an Intel processor, which TJ mentioned should probably work better than AMD for this stuff )....

 

Steve



I am using the free version of SPAMfighter for private users.
It has removed 459 spam emails to date.
Paying users do not have this message in their emails.
Try SPAMfighter for free now!
__._,_.___

Please note that this group is for discussion between users only.

To get support from AmiBroker please send an e-mail directly to
SUPPORT {at} amibroker.com

For NEW RELEASE ANNOUNCEMENTS and other news always check DEVLOG:
http://www.amibroker.com/devlog/

For other support material please check also:
http://www.amibroker.com/support.html




Your email settings: Individual Email|Traditional
Change settings via the Web (Yahoo! ID required)
Change settings via email: Switch delivery to Daily Digest | Switch to Fully Featured
Visit Your Group | Yahoo! Groups Terms of Use | Unsubscribe

__,_._,___