Message boards :
Number crunching :
Credits Proposal
Message board moderation
Previous · 1 · 2
Author | Message |
---|---|
mikey Send message Joined: 17 Dec 99 Posts: 4215 Credit: 3,474,603 RAC: 0 |
(snip)So if I understand you correctly you want each project to keep 100,000 units in storage at any one time just so they can average the time it took to process them? (snip) Since you put it that way I agree, the units ARE probably already stored. (snip).require a lot of processing power every day, every hour, every whatever, to then take all those times and average them?(snip) Yes, probably beyond current capabilities. The results are there, but the processing powere is not. [/quote] I still agree. |
Ingleside Send message Joined: 4 Feb 03 Posts: 1546 Credit: 15,832,022 RAC: 13 |
If only interested in the average, you don't need anything except total_cpu_time and number_of_results. When a new "success"-result is reported, just calculate: total_cpu_time = total_cpu_time + result_cpu_time number_of_results = number_of_results + 1 Whenever you wants to use average, just calculate total_cpu_time / number_of_results Whatever this average should be used for in credit-granting on the other hand is a mystery for me... |
Heffed Send message Joined: 19 Mar 02 Posts: 1856 Credit: 40,736 RAC: 0 |
We could also just do away with the credits and just crunch for the science. This option gets my vote. :) |
ML1 Send message Joined: 25 Nov 01 Posts: 20474 Credit: 7,508,002 RAC: 20 |
If only interested in the average, you don't need anything except total_cpu_time and number_of_results. Unfortunately, that 'average' is over simplistic. The variable nature of the WUs makes such an average worthless. When Astropulse and the Enhanced s@h client get into the mix, then the WU times will vary even more. Even for the present s@h WUs, you get a range of computation times depending on the angle range for the data in the WU (nicely explained here). You also get noisy WUs that can complete in just seconds! Thoughts for making the benchmarking a lot more accurate for each project for each host system is given on this thread (2nd: Credit for Clients). Try here for a summary. I don't give much credence to the 'credits'. However, for some, the credits are more highly valued than religion! And they add a competitive spirit ;) Regards, Martin See new freedom: Mageia Linux Take a look for yourself: Linux Format The Future is what We all make IT (GPLv3) |
Paul D. Buck Send message Joined: 19 Jul 00 Posts: 3898 Credit: 1,158,042 RAC: 0 |
Martin, Nice write up ... I added a link to it from the Wiki ... |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.