GeneralThis forum is for non-technical general discussion which can include both Linux and non-Linux topics. Have fun!
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
I was wondering if someone could tell me what kind of difference I could expect to see in processing speed doing algorithmic data crunching. I understand that for now using a 64 bit processor it will still only do the work at 32 bit. Does it however do the data crunching faster.? Added to that the dual core processor, does that make a difference.?
Is there a website that might explain this in further detail.
Thanks for any enlightment that can be shed on this subject.
Well... the first statement isn't necessarily true. There are native 64-bit versions of many Linux distributions now.
As far as performance... if your dealing with a threaded, CPU bound, application you could see a considerable performance increase (providing the program was well developed so the threads aren't locking each other all the time). If the application itself isn't threaded you will at least see more overall responsiveness while it is running.
If the process is I/O bound then your increase will be minimal.
Having messed around with a 4-way dual core Opteron (8 CPU's on 4 dies) I can tell you that running a parallel compile is ridiculously fast.
BTW... I'm moving this to General, as it isn't a Linux question.
Thanks for that info,that helps. How about something like running Seti, or Climateprediction. Will they see inproved preformance, and will my other programs still run at a good rate.?
I am using Libranet 2.8.1 now, but will shortly install Libranet 3.0
What kind of preformance gain will I actually see with this kind of system.? Am I stepping into a sports car when all I might need is a good solid and reliable family car.?
I know it might be hard to quantify, but some more info regarding the difference between my AMD 2800+ Barton, and the AMD Turion A64-Dual Core 3800+ running under Libranet 2.8.1/3.0 would help greatly.
This will not give you any performance increase for SETI or things similar to that, because they are not written in parallel. If you were to be writing the program for your algorithmic data crunching, you could write it to use both cores, otherwise it will run just like any other program would, using a single core. Granted, you will still have another core to use for encoding mp3s, SETI, etc. but those will not use both cores simultaneously. I've messed around w/ clustering a little bit, and it's the same concept, except on one die. It's also the same as an SMP machine, I believe