optimized applications and anonymous platform?
Message boards :
Number crunching :
optimized applications and anonymous platform?
Message board moderation
Author | Message |
---|---|
Send message Joined: 20 Jul 13 Posts: 15 Credit: 5,985,840 RAC: 0 |
greetings everyone, i just joined the project and noticed that on that A@H statistics page, some of the highest RACs belong to hosts w/ very few cores. for instance, the top host has a non-hyper-threading intel core i5 4-core/4-thread CPU and has a substantially higher RAC than the 2nd highest ranked host, which is an intel xeon 24-core host. then i realized that the core i5 host is completing its tasks 10 times faster than the intel xeon host, and that the xeon host is running the stock v100.00 application, while the core i5 host is running "anonymous platform." i searched the forum and came up empty-handed regarding info on any optimized applications and/or an associated app_info.xml file. obviously the project's top host platforms for which there are stock applications available are actually running an optimized application of some sort, or else their run times wouldn't be 10 times shorter than the average stock app v100.00 or v101.00 run time. could someone point me in the direction of the optimized app? even if it won't work with the one AMD host i've attached to the project with, i'd like to know the details in case i make the jump to an intel CPU. TIA, Eric |
Send message Joined: 1 Jun 13 Posts: 7 Credit: 0 RAC: 0 |
For example i5-3470 ~35K/day
|
Send message Joined: 26 Jan 13 Posts: 11 Credit: 13,585,797 RAC: 2,416 |
7 of the top 11 hosts are listed as "owner: anonymous". All are running "Period Search Application Anonymous platform (CPU)". All have quite amazing crunch times. This one: http://asteroidsathome.net/boinc/results.php?hostid=35494 shows crunch times more than ten fold faster than my 2600K. Is this some sort of closed beta testing of a new optimized application? |
Send message Joined: 9 Jun 12 Posts: 584 Credit: 52,667,664 RAC: 0 |
|
Send message Joined: 20 Jul 13 Posts: 15 Credit: 5,985,840 RAC: 0 |
No, I didn't send anyone my testing optimized apps. A few people ask me for source codes but I don't remember now who was it, so it is possible that those computers have some kind of optimized application. do you mean to say that the few people who you sent source code to may have compiled their own optimized apps? For example i5-3470 ~35K/day interesting that Free DC shows that host as having 8 CPUs, even though an i5 3470K is a non-hyper-threading quad core CPU...perhaps that host uses a dual-socket mobo? |
Send message Joined: 1 Jun 13 Posts: 7 Credit: 0 RAC: 0 |
We need to ask the user ..:: Thor ::.., this is his computers.
|
Send message Joined: 20 Jul 13 Posts: 15 Credit: 5,985,840 RAC: 0 |
We need to ask the user ..:: Thor ::.., this is his computers. i PMed him, so hopefully he'll get back to me via either PM or in this thread. |
Send message Joined: 18 Jun 12 Posts: 15 Credit: 5,027,400 RAC: 0 |
We need to ask the user ..:: Thor ::.., this is his computers. well - not a wild guess: PNT once again.. |
Send message Joined: 9 Jun 12 Posts: 584 Credit: 52,667,664 RAC: 0 |
Last modified: 23 Jul 2013, 8:12:45 UTC No, I didn't send anyone my testing optimized apps. A few people ask me for source codes but I don't remember now who was it, so it is possible that those computers have some kind of optimized application. Yes, they may have it. I will later release my optimized versions but I first have to test it, I have a new processor with AVX so I try a new optimization and I have to test that BOINC server will send the right version application to computer that is capable to compute with such instructions. |
Send message Joined: 1 Jun 13 Posts: 7 Credit: 0 RAC: 0 |
Yes, they may have it. I will later release my optimized versions but I first have to test it, I have a new processor with AVX so I try a new optimization and I have to test that BOINC server will send the right version application to computer that is capable to compute with such instructions. So put it on own server, we will help to test it via <anonymous platform>. |
Send message Joined: 26 Jan 13 Posts: 11 Credit: 13,585,797 RAC: 2,416 |
|
Send message Joined: 20 Jul 13 Posts: 15 Credit: 5,985,840 RAC: 0 |
I guess there are 2 important questions. Are the the computers running the optimized application producing scientifically valid results and if so, are the developer(s) willing to share the optimized source code with Kyong? i'm hoping that those who are currently utilizing optimized applications for this project (if that's in fact what's going on here) aren't PPD zealots. it will ultimately hurt the project if the people with the skills to create optimized apps kept those apps to themselves. of course i don't want to jump to conclusions and offend anyone, so perhaps no optimized apps have been published on the message boards yet b/c these developers haven't yet perfected their optimizations. for instance, while the project's top host (whose owner remains anonymous) has 5278 valids and 0 invalids, a substantial fraction of that host's errors are actual computation errors (as opposed to deadline lapses or server-side cancellations). i'm still waiting to hear back from Thor. |
Send message Joined: 20 Jul 13 Posts: 15 Credit: 5,985,840 RAC: 0 |
i got a response from Thor this morning... he is in fact using an optimized application, but he did not write it - the author of said application is a team member of his (he did not specify who). that user compiled and distributed the optimized app to his teammates for testing purposes just before leaving on vacation. its not real clear whether the app is still being tested or if its ready for release...and i don't know when the author is due back in town, but Thor knows to get back to me once he knows more. |
Send message Joined: 18 Jun 12 Posts: 15 Credit: 5,027,400 RAC: 0 |
i got a response from Thor this morning... ask sesef.. |
Send message Joined: 26 Jan 13 Posts: 11 Credit: 13,585,797 RAC: 2,416 |
I guess there are 2 important questions. Are the the computers running the optimized application producing scientifically valid results and if so, are the developer(s) willing to share the optimized source code with Kyong? Only Kyong can confirm whether or not the results the optimized app are producing are scientifically valid or not. I don't trust boinc to declare the results to be valid. I hope I'm wrong though. |
Send message Joined: 1 Apr 13 Posts: 37 Credit: 153,496,537 RAC: 0 |
Last modified: 3 Aug 2013, 22:16:54 UTC It would be really nice if we could get nearly as well optimized applications like this: http://asteroidsathome.net/boinc/results.php?hostid=37108 To 8 times faster than my shortest periods (~1200 sec vs. ~9800 sec)! |
Send message Joined: 9 Jun 12 Posts: 584 Credit: 52,667,664 RAC: 0 |
|
Send message Joined: 21 Dec 12 Posts: 176 Credit: 136,462,135 RAC: 0 |
It would be really nice if we could get nearly as well optimized applications like this: http://asteroidsathome.net/boinc/results.php?hostid=37108 We are working hard. :-) |
Send message Joined: 20 Jul 13 Posts: 15 Credit: 5,985,840 RAC: 0 |
this is great news! it appears that there are multiple optimized apps being developed right now - are multiple developers competing to create the most efficient optimized app for a single platform? or has the work been split up such that an optimized app is being developed as an alternative to each of the 11 standard apps? |
Send message Joined: 21 Dec 12 Posts: 176 Credit: 136,462,135 RAC: 0 |
|
Message boards :
Number crunching :
optimized applications and anonymous platform?