- cmsg cancel <nd9029$14v$2@dont-email.me> - 5 Updates
- Please read more the following about USL... - 1 Update
- Why i have posted my USL or Delphi and FreePascal in this group ? - 1 Update
- More about my USL for Delphi and FreePascal... - 1 Update
- I correct, please read again... - 1 Update
- And here is important thing about a USL benchmark.. - 1 Update
| bleachbot <bleachbot@httrack.com>: Mar 27 06:03PM +0200 |
| bleachbot <bleachbot@httrack.com>: Mar 27 06:11PM +0200 |
| bleachbot <bleachbot@httrack.com>: Mar 27 11:12PM +0200 |
| bleachbot <bleachbot@httrack.com>: Mar 27 11:50PM +0200 |
| bleachbot <bleachbot@httrack.com>: Mar 28 12:13AM +0200 |
| Ramine <ramine@1.1>: Mar 27 06:17PM -0700 Hello.... Please read more the following about USL... I have just included a windows executable called usl.exe of my USL for Delphi and FreePascal inside the zip to easy the job for you... Please read the readme file inside the zip to understand how to use it... So hope that you will be happy with it.. It is a very important and very powerful tool that predict scalability and that allow you also to optimize more efficiently the criteria of cost of your multicores systems etc. You can download the windows executable and the source code of my USL for Delphi and FreePascal from: https://sites.google.com/site/aminer68/universal-scalability-law-for-delphi-and-freepascal Thank you, Amine Moulay Ramdane. |
| Ramine <ramine@1.1>: Mar 27 05:54PM -0700 Hello... Why i have posted my USL or Delphi and FreePascal in this group ? Because it is a tool as a program that you must compile with FreePascal into an executable without the need to download and learn R language, because the only other tool that you will find that do my USL for Delphi and FreePascal job is this R package: http://www.inside-r.org/packages/cran/usl/docs So i have brought you a second and new tool that do the job without needing to install the R statistical software. So please download my software here: https://sites.google.com/site/aminer68/universal-scalability-law-for-delphi-and-freepascal And download FreePascal to compile it as an executable from here: http://www.freepascal.org/ Thank you, Amine Moulay Ramdane. |
| Ramine <ramine@1.1>: Mar 27 05:16PM -0700 Hello, More about my USL for Delphi and FreePascal... I have just tested more my USL for Delphi and FreePascal and it is working perfectly. But to make it work best with multicores, you have to choose the first column of the number of cores of the csv to: 1,2,4,8,16 without going up to 32 and it will work ok, don't choose 1,2,3,4,5 because this will not be a good approximation and this will cause problems. You can download my USL for Delphi and FreePascal from: https://sites.google.com/site/aminer68/universal-scalability-law-for-delphi-and-freepascal Thank you, Amine Moulay Ramdane. |
| Ramine <ramine@1.1>: Mar 27 12:14PM -0700 Hello... I correct, please read again... Hope you have read my previous post.. And here is an important thing about a USL benchmark.. Read here: https://www.vividcortex.com/blog/2015/11/13/capacity-planning-with-usl/ And read about the benchmark: https://www.percona.com/docs/wiki/benchmark:cisco:scale:start And here is the server configuration: https://www.percona.com/docs/wiki/benchmark:hardware:cisco_ucs_c250 As you have noticed this guy have used USL to predict scalability, but i don't think he is right, because the performance data comes from also hyper-threading, because the performance data goes up to 32 threads, so because there is a mix of hyper-threads and threads, so this have caused USL to not predict correctly the peak scalability etc. because the regression can not extract the right information in this scenario, so you have for example to benchmark with up to 12 thread to avoid hyper-threading. Thank you, Amine Moulay Ramdane. |
| Ramine <ramine@1.1>: Mar 27 12:06PM -0700 Hello, Hope you have read my previous post.. And here is important thing about a USL benchmark.. Read here: https://www.vividcortex.com/blog/2015/11/13/capacity-planning-with-usl/ And read about the benchmark: https://www.percona.com/docs/wiki/benchmark:cisco:scale:start And here is the server configuration: https://www.percona.com/docs/wiki/benchmark:hardware:cisco_ucs_c250 As you have noticed this guy have used USL to predict scalability, but i don't thing he is right, because the performance data comes from also hyper-threading, because the performance data goes up to 32 threads, so because there is a mix of hyper-threads and threads, so this have caused USL to not predict correctly the peak scalability etc. because the regression can not extract the right information in this scenario, so you have for example to benchmark with up to 12 thread to avoid hyper-threading. Thank you, Amine Moulay Ramdane. |
| You received this digest because you're subscribed to updates for this group. You can change your settings on the group membership page. To unsubscribe from this group and stop receiving emails from it send an email to comp.programming.threads+unsubscribe@googlegroups.com. |
No comments:
Post a Comment