The system for folding statistics from Stanford (or is it at PG) is a disaster, having an average total collapse every other month. The last time it happened was just over a month ago, between July 11 to 14, and it occurs almost invariably in a weekend where it is impossible to make contact with staff who can sort out the problems. If I had had such high downtime of my servers, I had resigned myself from the job. And I'm getting pretty tired of seeing that things are not working as they should, so maybe Stanford should outsource this job to someone who can get it to work?
I guess this post will be deleted quite fast, and that I might get banned for criticizing the "who" that is responsible for the statistics to work. However, my criticism is not unreasonable, since the same collapse repeats itself time and time again.
I don't know how long you've been folding but these things happen. All distribted computing projects have problems now and then - some more then others. If the science gets done, then you've done your job. The stats are just there to tell your grandkids what you did in the ancient past - and anoy your team members who only have old crappy PCs.
John_Weatherman wrote:I don't know how long you've been folding but these things happen. All distribted computing projects have problems now and then - some more then others. If the science gets done, then you've done your job. The stats are just there to tell your grandkids what you did in the ancient past - and anoy your team members who only have old crappy PCs.
I started with folding in november 2008, and I believe I have given my share to science, just check the stat by http://folding.extremeoverclocking.com/ ... ist.php?s=
The stats problems never really bothered me that much. No points/work is lost and they will catch up. I've worked with computers enough to know that even a well built computer will find a way to crash when I'm away from home. And I imagine the current stats server isn't cutting edge - that isn't really where they should be spending a bunch of money (as the current one usually work fine). That being said, I've noticed recently that PG seems to be 'on the ball' with server issues (getting them fixed quickly over the weekend). So good job PG - just wanted to let you know it is appreciated.
And as a another pointed out, I've had more issues with BOINC servers in general (though some projects are much better than others with reliability and communication).
....
