| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
| World Community Grid Forums
|
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 99
|
|
| Author |
|
|
KerSamson
Master Cruncher Switzerland Joined: Jan 29, 2007 Post Count: 1684 Status: Offline Project Badges:
|
Hi Speedy51,
----------------------------------------the question is not directly related to the granted points but to the computational efficiency. 40 CPU hours = 40 hours of electricity power (not for free) If HST1 grants so few credits, my conclusion is that the project is wasting CPU resources instead to be efficient. Such conclusion causes my observations since a couple of months. If the CPU power is misused by a project it is not fair for the other projects regarding the wasted CPU time, and it is not correct regarding the contributor's electricity bill. At this time, I do not feel that the HST1 scientists are really aware/listening about the multiple issues caused by the HST1 science. I am only expecting that someone would say: "I understand the concern and the frustration, we will investigate for better efficiency and less troubles, and we will keep you in touch". Since I am contributing to WCG only, I don't really care of credits comparing other grid projects. However, I try to optimize the delivery of computational resources by my hosts ... based on the granted credits, since we do not have any other "efficiency measurements". Cheers, Yves |
||
|
|
Speedy51
Veteran Cruncher New Zealand Joined: Nov 4, 2005 Post Count: 1326 Status: Offline Project Badges:
|
Hi Yves,
----------------------------------------I can kind of understand where you're coming from in regards to the credit. I hope you get the outcome you are after For me personally it is more getting the work processed & run-time for badges as opposed to credit. ![]() [Edit 1 times, last edit by Speedy51 at Sep 28, 2016 7:26:59 AM] |
||
|
|
KerSamson
Master Cruncher Switzerland Joined: Jan 29, 2007 Post Count: 1684 Status: Offline Project Badges:
|
As expected, WU HST1_007888_000051_KT0001_T350_F00018_S00006 (39+ hours) invalid as well.
----------------------------------------Yves --- @Speedy51: I am not sure to really understand your remark. My current credit reflects only 9.5 years contribution with a couple of hosts (currently 6 machines, not the most modern one). |
||
|
|
Speedy51
Veteran Cruncher New Zealand Joined: Nov 4, 2005 Post Count: 1326 Status: Offline Project Badges:
|
@Speedy51: I am not sure to really understand your remark. My current credit reflects only 9.5 years contribution with a couple of hosts (currently 6 machines, not the most modern one). I don't take a lot of notice about the points. I am more interested in run time because it gains you badges. I have edited the post I wrote on the 26th. I think you will find it much easier to read ![]() [Edit 1 times, last edit by Speedy51 at Sep 28, 2016 7:29:54 AM] |
||
|
|
KerSamson
Master Cruncher Switzerland Joined: Jan 29, 2007 Post Count: 1684 Status: Offline Project Badges:
|
Hi Speedy51
----------------------------------------indeed I am less focused on time since I am especially looking of crunching efficiency. Amongst others, my background is to monitor processes. I use a similar approach to my WCG contribution. For me the overall amount of points is not particularly relevant, but the computation efficiency of each host. The points are the only available measurement for doing it. Cheers, Yves |
||
|
|
armstrdj
Former World Community Grid Tech Joined: Oct 21, 2004 Post Count: 695 Status: Offline Project Badges:
|
Yves,
Sorry to hear you are still getting invalids. With the latest updates to the applications we have seen great improvement to the invalid rate but there are still a couple of things we are looking at to help more. For the credit we have made changes some time ago to help with this and things have improved. The credit awarded is not a good indication of cpu efficiency. For this project there are two types of tasks with a large variation in cpu time for those. This is expected. Thanks, armstrdj |
||
|
|
KerSamson
Master Cruncher Switzerland Joined: Jan 29, 2007 Post Count: 1684 Status: Offline Project Badges:
|
Hi amstrdj,
----------------------------------------I thank you warmly for your feedback. I was business traveling last week without any ability to monitor my machines accurately. As I can see during the last couple of days, it looks like a little bit better in terms of "invalid" WUs. Cheers, Yves |
||
|
|
KerSamson
Master Cruncher Switzerland Joined: Jan 29, 2007 Post Count: 1684 Status: Offline Project Badges:
|
I have an additional remark.
----------------------------------------8 days ago, the concerned machines - Ubuntu 14.04 x64 - moved from kernel version 3.13.x to 4.4.x. Does it impact the better validity of the HST1 results? Yves |
||
|
|
armstrdj
Former World Community Grid Tech Joined: Oct 21, 2004 Post Count: 695 Status: Offline Project Badges:
|
No the kernel version should not have an effect one way or the other.
Thanks, armstrdj |
||
|
|
|