List Archives >  Mathcad List Archive >  Archive by date >  This Month By Date >  This Month By Topic

[mathcad] Re: Multiprocessor system

Search email archive for  

[mathcad] Re: Multiprocessor system
Author: Colin G Clarke    Posted: Mon, 28 Jul 2003 10:15:51 +0000
Unless anyone knows different, I think the simple answer is: On a UNIX system - Yes - The OS automatically takes advantage of multiple processors. On a MS Windows system - No - Unless the software is specifically written for multiple processors, there will be no benefit. I don't believe that MathCad has a multiprocessor version. Regards, Colin Clarke. Zdenek Sedlacek on 28/07/2003 08:56:35 Please respond to /> To: Mathcad Discussion List /> cc: Fax to: Subject: [mathcad] Multiprocessor system ************* Message inserted by MIMEsweeper *********** This email, received by bnfl.com has been virus scanned by MIMEsweeper using DrSolomons TVD. Please treat all files received via the internet with extreme caution. ******************************************************************** Hello Mathcad users! Could anyone please tell me whether there is any sense in running Mathcad on a multiprocessor system? Thank you. Zdenek Sedlacek ------------------------------------------------------------------ Dr Zdenek Sedlacek Institute of Plasma Physics phone: ++420-266 052 852 Academy of Sciences fax: ++420-286 586 389 P. O. Box 17 telex: 122018 atom c Za Slovankou 3 e-mail: /> CZ-182 21 Prague 8 Czech Republic home: ++420-251 641 487 ------------------------------------------------------------------ --- The Mathcad List - Discussion, Support & News Contributions: /> Hosted by: Adept Scientific http://www.adeptscience.com List Archive: http://lists.adeptscience.co.uk/ --- Would you like this to come to a different email address? --- Simply leave the mailing list (see below) and re-join by sending a blank email from the new address to: or --- If you do not want to be on the Mathcad list, simply send a blank email (no subject or message needed) to: /> The information contained in this email may be commercially sensitive and/or legally privileged. It is intended solely for the person(s) to whom it is addressed. If you are not a named recipient, you are on notice of its status. Please notify the sender immediately by reply e-mail and then delete this message from your system. You must not disclose it to any other person, copy or distribute it or use it for any purpose. --- The Mathcad List - Discussion, Support & News Contributions: /> Hosted by: Adept Scientific http://www.adeptscience.com List Archive: http://lists.adeptscience.co.uk/ --- Would you like this to come to a different email address? --- Simply leave the mailing list (see below) and re-join by sending a blank email from the new address to: or --- If you do not want to be on the Mathcad list, simply send a blank email (no subject or message needed) to: />

[mathcad] Re: Multiprocessor system
Author: Quang-Viet Nguyen    Posted: Mon, 28 Jul 2003 10:44:07 -0400
On an N-processor system, you can run N-instances of mcad simultaneously without a loss of performance. This is useful for parametric studies where you will have a lot of runs. However, with only one mcad instance you will not get any speed advantage. With some programming effort, one could develop a parallel processing dll that will invoke multi-processors in a networked cluster or multi-CPU SMP environment. This would be one way to get true multi-cpu performance speedup for a single mcad problem. Quang-Viet Nguyen At 10:56 AM 7/28/2003 +0200, you (Zdenek Sedlacek) wrote: >Hello Mathcad users! > >Could anyone please tell me whether there is any sense in running Mathcad >on a multiprocessor system? >Thank you. > >Zdenek Sedlacek >------------------------------------------------------------------ >Dr Zdenek Sedlacek >Institute of Plasma Physics phone: ++420-266 052 852 >Academy of Sciences fax: ++420-286 586 389 >P. O. Box 17 telex: 122018 atom c >Za Slovankou 3 e-mail: /> >CZ-182 21 Prague 8 >Czech Republic home: ++420-251 641 487 --------------------------------------------------------------- Quang-Viet Nguyen, Ph.D. NASA Glenn Research Center 21000 Brookpark Rd, MS 5-10 Cleveland, Ohio 44135 Voice: (216) 433-3574 Fax: (216) 433-5802 E-Mail: (best way to reach me) --------------------------------------------------------------- --- The Mathcad List - Discussion, Support & News Contributions: /> Hosted by: Adept Scientific http://www.adeptscience.com List Archive: http://lists.adeptscience.co.uk/ --- Would you like this to come to a different email address? --- Simply leave the mailing list (see below) and re-join by sending a blank email from the new address to: or --- If you do not want to be on the Mathcad list, simply send a blank email (no subject or message needed) to: />

[mathcad] RE: Multiprocessor system
Author: Dave Robinson    Posted: Sat, 2 Aug 2003 15:45:34 +0100
I have installed two different versions of MathCad onto dual processor Pentium systems. The first was on an NT 4.0 box using a MathCad 8 system with all its service packs included. The setup was completely unstable, with regular monotony, and without any warning MathCad, would just cannon itself out of the system leaving you looking at your blank desktop. The only way we got some improvement was to turn one of the processors off. I abandoned that machine and went back to a higher performance single processor computer, where MathCad run as well as version 8 ever did. I currently have MathCad 2001i installed on a dual Pentium system at work running under Windows 2000, I have also got it installed on my company laptop, which is a high performance Dell Inspiron desktop replacement unit. On the latter machine, it runs like a dream, on the dual processor machine it runs like a pig, again occasionally (but rarely) cannoning out of the sytem. As you would expect, when only running MathCad, only 50% of my computing resources are used; however this often shoots up to 100% with the machine in some form of limbo state. This can last up to several minutes, before it recovers. Despite my dual processor being a slightly higher 'spec'ed machine than the laptop, the laptop will invariable undertake the calculation faster than the dual processor machine. My advice - for what it is worth, go for the fastest single processor machine you can afford. Regards Dave Robinson -----Original Message----- From: Zdenek Sedlacek Sent: 28 July 2003 09:57 To: Mathcad Discussion List Subject: [mathcad] Multiprocessor system Hello Mathcad users! Could anyone please tell me whether there is any sense in running Mathcad on a multiprocessor system? Thank you. Zdenek Sedlacek ------------------------------------------------------------------ Dr Zdenek Sedlacek Institute of Plasma Physics phone: ++420-266 052 852 Academy of Sciences fax: ++420-286 586 389 P. O. Box 17 telex: 122018 atom c Za Slovankou 3 e-mail: /> CZ-182 21 Prague 8 Czech Republic home: ++420-251 641 487 ------------------------------------------------------------------ --- The Mathcad List - Discussion, Support & News Contributions: /> Hosted by: Adept Scientific http://www.adeptscience.com List Archive: http://lists.adeptscience.co.uk/ --- Would you like this to come to a different email address? --- Simply leave the mailing list (see below) and re-join by sending a blank email from the new address to: or --- If you do not want to be on the Mathcad list, simply send a blank email (no subject or message needed) to: /> --- The Mathcad List - Discussion, Support & News Contributions: /> Hosted by: Adept Scientific http://www.adeptscience.com List Archive: http://lists.adeptscience.co.uk/ --- Would you like this to come to a different email address? --- Simply leave the mailing list (see below) and re-join by sending a blank email from the new address to: or --- If you do not want to be on the Mathcad list, simply send a blank email (no subject or message needed) to: />

Previous by date: [mathcad] Multiprocessor system, Zdenek Sedlacek
Next by date: [mathcad] Re: 6 decimal place precision,  Richard Jackson
Previous thread: [mathcad] Re: 6 decimal place precision,  Sven Lindhardt
Next thread: [mathcad] Re: 6 decimal place precision,  Sven Lindhardt