Posts by NJMHoffmann

1) Message boards : Current tests : New crediting system (Message 2037) Posted 13 Aug 2006 by NJMHoffmann Post: Here's the latest. Hope it's visible as it was shrunk a great deal. All results are from boinc 5.5.11 (standard). Why so many statistics with this 2 cr/model? Ralph is testing the software, the credits are nonsense now. They tell us nothing, because Rosetta will have different fixed credits/model for each WU-type. What might be interesting is, if the times/model for a given WU-type are consistent enough to go to a fixed credit model. Norbert
2) Message boards : Current tests : New crediting system (Message 2017) Posted 12 Aug 2006 by NJMHoffmann Post: For R@h, we will use the average value from the Ralph runs so everyone will get the same credit/model for a given work unit rather than a value that may change a bit initially. I propose not only using Ralph-values for Rosetta, but to include the values received so far from Rosetta (per WU-type ) too. You'll need less Ralph units of a new WU-type to start on Rosetta, because the credit per WU granted from Ralph-estimations is only a starting point, that will be calibrated further with the returned Rosetta-WUs. Norbert
3) Message boards : RALPH@home bug list : Host venues corrupted? (Message 1949) Posted 10 Aug 2006 by NJMHoffmann Post: With the last contact with ralph@home all my hosts logged: New host venue: <some numerical value>. The "my hosts" info looks ok. Norbert
4) Message boards : RALPH@home bug list : Bug reports for Ralph 5.21 (Message 1784) Posted 6 Jun 2006 by NJMHoffmann Post: For this WU for example it took 25 minutes between the checkpoints (models) which can translate in over an hour on a slow Mac. And this was one of the small (t314 with Nres=106) targets. Over at Rosetta people are "complaining" that it may take between 90-120 minutes for a WU to reach its first checkpoint. I am running two WUs for the slightly bigger targets 299 (Nres=180). One is at 40 min., one at 35. No checkpoint until now. (Edit: finished first models at 44/38 mins.) You can imagine how the t296 with Nres=445 looked like. What happened to more often checkpointing? Inquiring minds want to know :-) Norbert (waiting for the boinc client, that waits for a checkpoint before switching the task)
5) Message boards : RALPH@home bug list : Bug reports for Ralph 5.20 (Message 1766) Posted 4 Jun 2006 by NJMHoffmann Post: How long before you should abort WU's stuck at 100%? Why does my firewall show a lot of traffic for bonic ralph client even though its stuck at 100% I have all other projects suspended to see if the WU will report. Wild guess: The client is downloading (BIIIG) symbol tables for the debug output?? Norbert
6) Message boards : RALPH@home bug list : Target runtime for Ralph (Message 563) Posted 24 Feb 2006 by NJMHoffmann Post: The preferences page tells, that "not selected" defaults to a target runtime of 1 hour. The progress of my 4.87 WUs suggest, that the default is 8 hours like in Rosetta. Norbert
7) Message boards : Number crunching : No work from project. (Message 418) Posted 21 Feb 2006 by NJMHoffmann Post: There is a bug IMHO in the BOINC client that leeds to a request for far too much work for low ressource projects (esp. after a project ran "dry" of WUs). Is it possible to change the server for RALPH so that only one WU is distributed, ignoring the seconds the client asked for? This would prevent those "large cache by accident" and would give you shorter turnaround times. Norbert Norbert: Actually it is not a bug it is just part of the production environment that is not compatible with a testing environment. While your point is well taken and the suggestion is a good one, It might not be so simple to implement. I will send the idea along to the project. If you have a host, that crunches e.g. for Rosetta 90% and for Ralph 10%, if there is new work for Ralph it will ask for 86400 seconds of new work (at a setting of 1 day connect time). The client or the server would have to make that 8640 seconds (= 10% of a day). Now both parts rely on the other side to do this calculation, but it is done nowhere. I'd call it a bug. Norbert
8) Message boards : Number crunching : No work from project. (Message 334) Posted 19 Feb 2006 by NJMHoffmann Post: THIS IS A TEST PROJECT. THERE WILL BE NO WORK UNITS AVAILABLE AT TIMES. THEY GET ISSUED WHEN NECESSARY FOR TESTING. PLEASE SET THE RESOURCE SHARE LOW FOR THIS TEST PROJECT SO PRODUCTION PROJECTS GET HIGHER PRIORITY This project is not for science, credits, or other competitive statistics, but only for testing to make our production project, Rosetta@home, better. There is a bug IMHO in the BOINC client that leeds to a request for far too much work for low ressource projects (esp. after a project ran "dry" of WUs). Is it possible to change the server for RALPH so that only one WU is distributed, ignoring the seconds the client asked for? This would prevent those "large cache by accident" and would give you shorter turnaround times. Norbert