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: 33
|
![]() |
Author |
|
Viktors
Former World Community Grid Tech Joined: Sep 20, 2004 Post Count: 653 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
I have a script on the servers which estimates, once per week, how far along we are on the project. There is a lot of variability in the difficulty of the work and in contribution rate, so it is hard to give a precise number. However, the script currently says we are somewhere between 55% and 67% done. Hence, we say we are over 60% done to simplify things.
|
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I had a quick look at the global statistics. It seems as though around the first week of March we peaked out with the maximum numbers of results posted and the maximum number of points generated.
----------------------------------------Ever since then, there has been a downward trend on both types of data. If someone has all the raw data ( I believe Christine mentioned that), then a quick spread sheet dump of some accumulated data (without details) will help identify the source of the problem (the statistical data on the web site is text so you cannot work with it unless some is clever enough to convert it or patient enough to reenter it). Suggest calculating and putting some trends onto graphs: 1. active new users 2. avrg points generated by new members ( new member can be defined up to one month old) 3. users older than one month that have stopped being active 4. avrg points returned by active users 5. total number of active users This will give us our “churn” in terms of users who are no longer producing points to allow us to see what is pulling us down. These maybe old users or new users. We need to assess the number of new users that drop off quickly. This is often a problem with mass services - they have an unfulfilled expectation and leave quickly. Since they are new, they have no loyalty to the service and don't try to resolve the problems they encounter. With all that being said, hats off to the person that writes the scripts to get the data out of the terabyte databases !!! I would be very cautious about speculating on the actual cause of the change in trend as it terribly difficult to be accurate without getting into the details of the data. Eg. Loosing someone who has been generating a couple of tens of thousands of points and loosing someone who signed up for a week is a totally different problem and needs to be addressed differently. I can work with someone to analyze the data as the definitions of who is active and who is no longer active is critical to the analysis – I do have expertise in this field – please contact me directly on xxxx@yyyyy.zz [post edited to remove email address at users request - knreed] [Edit 1 times, last edit by knreed at Aug 1, 2005 1:56:00 PM] |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I believe Christine mentioned that sorry it was Viktors who made that staement - I appologise Christine. RobertS |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Cool.. That means we should be done around September / October something.. would be great!
|
||
|
Kdemonn
Cruncher Joined: Nov 16, 2004 Post Count: 22 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Actually, I am basing this on us continuing at the average rate we were going at over the past month, not on our rate since the project started. I was a bit disappointed to see that the project actually does not seem to be picking up momentum. If anything we seem to be slowing down. If you don't believe, check out the global statistics history. In April, it was really common for the grid to reach over 60 years in processing time, but that level hasn't been reached at all in May. It's also been weeks since we reached more than 13 million points on any given day, a number we used to reach routinely. Does anyone have an idea what is going on and what can be done to turn things around? It seems like numbers of members is increasing, but people are not contributing as much as they used to. Sorry to be the bearer of bad news. I can't speak for anyone else, but in my case over the winter (I live in Michigan) I was running six machines 24/7 for the project. This kept my home office and environs relatively warm, and greatly reduced the amount the furnace ran. In essense, I was only paying the difference between the cost of electric heat and gas heat, in order to contribute to scientific advancement. Now that it's summer, however, I have to pay the full cost of the electricity for the computers, plus additional charges to run the air conditioner to remove excess heat. Hence, I'll only be running the more powerful machines, as they give the best computational return per watt used. (And even they get turned off during thunderstorms.) |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I must also admit that I have had to slow my computers down due to
heat |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Kdemonn from Michigan: I concur with your seasonal use patterns as Nebraska has similar issues; especially those thunderstorms that fry the silly-cone. I have seen some nasty cooking jobs after a blast of light that left owners without any backups of lots of work.
![]() ![]() My basement is HQ so the heat in winter is welcome, but even in summer it has little impact on cooling. Spent the past 25 years insulating and rebuilding the bricks & boards. About 24" of attic insulation has been a money savings that the local electric shop did not get. |
||
|
David Autumns
Ace Cruncher UK Joined: Nov 16, 2004 Post Count: 11062 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Hi I was thinking about this as I was studying the points per day trend which peaked at ~14,000,000 and we are now running at about 11,000,000 a day.
----------------------------------------I thought maybe it's because it's summer and people are out and about taking in the fresh air as these computer things are always much more interesting when it's dark and cold outside and the warm air eminating from your PC is quite comforting against a chilly evening and then I realised..... This is the World Community Grid and it's only Summer for half the Planet ![]() Now might be the time to publicise the Grid in the Southern Hemisphere ![]() Keep 'em crunching Dave ![]() |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Now might be the time to publicise the Grid in the Southern Hemisphere ... you stole my thoughts.... let's get those invitations out to the south.... and maybe you great team bashers can start competing in the International League, now that you have your successes North American one ![]() ![]() ![]() |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
We can't forget that some over at Grid.org are also crunching on HPF. That should figure into the calculation. They run the same Rosetta software we do. Plus, it looks like they will be installing a project status side bar. Let's hope we can get one over here too. I left Grid.org for WCG because I think it's more user-friendly for a computer plebeian like myself and I just wanted to work on HPF. I like it here better.
|
||
|
|
![]() |