Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
World Community Grid Forums
Category: Support Forum: Suggestions / Feedback Thread: [Solved] WCGrid Cost-Benefit Analysis |
No member browsing this thread |
Thread Status: Active Total posts in this thread: 133
|
Author |
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
If you are waiting on a cost benefit statement from the WCG group, I can say with almost certainty you are not going to get it. They have a large to-do list that has been around for years that they haven't worked on yet plus they are trying to get some climate projects on-boarded. I would prefer they work on those tasks versus some nebulous cost-benefit analysis that, in the end, is of little to no value. If you think the benefit doesn't out tway the cost then stop contributing. We each make our own value decision on whether to contribute or not.
|
||
|
cowtipperbs
Advanced Cruncher Joined: Aug 24, 2009 Post Count: 78 Status: Offline Project Badges: |
wcgridmember -
----------------------------------------Take a look at the BOINC Workshop 2019 that was from July 9th - 12th https://docs.google.com/document/d/1pH-3q4iEU...eXqncYD7T2s-XngrYhIc/edit . On Day 2 July 10th 1:00 they had a talk on "The heating effect and the power consumption of running BOINC on a desktop (Andy Bowery, University of Oxford)." You can view the stream here: https://www.youtube.com/channel/UCJDOd7pQf1IF8ZxKHpNwung. Don't know if this will help out or not |
||
|
wcgridmember
Advanced Cruncher Joined: Mar 30, 2005 Post Count: 110 Status: Offline Project Badges: |
Thank you for your contributions! I will definitely watch that stream.
|
||
|
aaabaaab
Cruncher Joined: Jul 31, 2016 Post Count: 42 Status: Offline Project Badges: |
Could it not be the case that you are sent less computing intensive work units to your lower end devices? It could be but I don't know how to get more details on this... |
||
|
wcgridmember
Advanced Cruncher Joined: Mar 30, 2005 Post Count: 110 Status: Offline Project Badges: |
Hi! I guess if WCGrid developers are not willing to try and make a point of a cost-benefit analysis - such as the more minimalist version I suggested: if the world gets these or those results in X decades, all our CO2 emissions made until then will have been worthwhile with a great amount of certainty, say 95% (X could be increased as much as needed to attain the 95% certainty) - maybe they think that's too early a point to make, then they could at least make a point of costs alone - WCGrid members together are emitting around Y tons of CO2 per year (based on the number of users per country, the energy sources breakdown per country and an estimation of the power consumptions), Z per active device or thread (at 60% of CPU usage) on average. This is easier than making a cost-benefit analysis and would go a long way in giving WCGrid much more transparency and credibility.
----------------------------------------[Edit 2 times, last edit by wcgridmember at Jul 23, 2019 1:44:53 PM] |
||
|
wcgridmember
Advanced Cruncher Joined: Mar 30, 2005 Post Count: 110 Status: Offline Project Badges: |
Hi, @aaabaaab. Maybe you could adjust those values to the points obtained for the tasks your devices are completing.
|
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Could it not be the case that you are sent less computing intensive work units to your lower end devices? It could be but I don't know how to get more details on this... Absolutely, positively not. This proposition never got beyond a paper which I think the father of BOINC (co)wrote circa previous decade. IIRC, there are comments somewhere by knreed on these forums about this. Rosetta had/has a profile function by which one could specify how long a task would run, something like 2-4-6-8 hours. By that token it hardly mattered how powerful a device is. Whether a simulation has equal scientific value at 2 hours opposed to one running 8 hours, ask the scientists. |
||
|
wcgridmember
Advanced Cruncher Joined: Mar 30, 2005 Post Count: 110 Status: Offline Project Badges: |
Maybe another way to approach this problem would be to use the measure "points/Wh" (on average, on a reasonable time-scale), no?
|
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
No. As has been talked about ad nausem in these forums, the points calculations are not accurate. A worthless statistic...
|
||
|
wcgridmember
Advanced Cruncher Joined: Mar 30, 2005 Post Count: 110 Status: Offline Project Badges: |
Thanks! I had no idea before your post. I'm sorry for not taking more due diligence to search that out.
|
||
|
|