I have been running climatePrediction under BOINC since 11 March. The BOINC manager shows zero credit. The \"my account\" page shows the same.

The BOINC manager shows that climatePrediction *is* running and that it has used 35:13:36 hours of CPU

On the other hand, I also run SETI and Einstein under BOINC on the same machine (Windows XP/SP2). Both these show positive credit.

What is wrong please?

My guess would be that the climateprediction task is not running for long enough, and may not even be checkpointing (reaching a save point).

Credit is awarded on trickles to the server which occur at the end of each model year (1 Dec). You can see how far the model has got in the graphics window of the work (tasks) tab. The first is due at 1 Dec 1921.

The checkpoints are set at the end of each 6 model days (00.30 on the 1st, 7th, 13th of each model month). There is a countdown to the checkpoint in the graphics window with key 8 (Z to remove the side panel). If you have preferences set to remove the model from memory when preempted, or it is lost to memory eg when you shut down, then progress made since the last checkpoint will be lost.

---

The PC has only 256mb, which is below the recommended minimum (512mb), thus may be running very slowly. The \'%\' complete would be useful...

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

The PC has only 256mb, which is below the recommended minimum (512mb), thus may be running very slowly. The \'%\' complete would be useful...

I changed the options in the light of the previous poster\'s comments about preemption (thanks for that) but it made my system completely unstable. I believe there is a quality defect in climatePrediction which causes it not to run according to preferences; I had it set to run only when the machine is idle but it was constantly using 60-70% CPU.

Thank you anyway.

Running at 60-70% CPU rather than 100% is a sign of a memory shortage. The disk will be going bonkers, and it\'ll also be very slow to switch between tasks (sometimes 10 minutes or higher).

Because it takes so long to page stuff into and out of memory, by the time the computer has noticed you\'re not idle, started to page the model out of memory, it takes long enough so it thinks you are idle again.

Increasing memory would help enormously.

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

Thanks. I\'ll think about getting some more memory. Not sure what the expandability is like on this machine. In the meantime I\'m afraid I\'ve completely removed climatePrediction from the system - SETI and Einstein don\'t seem to suffer from these problems; presumably they don\'t need as much memory for their calculations.

Yes, the coupled model takes 90mb, I think Seti is about 50?

---

I'm a volunteer and my views are my own.
News and Announcements and FAQ

That\'s not a problem with this project per se... That\'s more an issue that your machine isn\'t up to spec. SETI and Einstein have less stringent resource requirements, that\'s why you don\'t see that.

When a project puts recommendations on their site for a given model, much like a software company puts it on the box, there is a certain degree of reasoning behind those recommendations.

One other thing of note... Having things set to \"leave in memory\" can help wrt the not reaching a checkpoint issue. It is also a normal recommendation for a project such as this. However, and here\'s the unfortunate part, \"leave in memory\", has this stuff paged out, and if you\'re having to page a hell of a lot already, hmm... You\'re committed memory between physical and pagefile will definitely go up keeping all projects in memory...

---

I am having a similar problem: I have been running a climateprediction model since 28 April. I have had 7 trickles, and my model is now in August 1927. I have no credits since I started this model. Any idea what\'s the problem?

---

The new trickle server is still trying to catch up with the back log.

The new trickle server is still trying to catch up with the back log.

I don\'t think it\'s catching up. The project\'s dayly credits decreased from an average of 3.0M (2 weeks ago) to less than 0.7M yesterday. See Boincstats
This means the backlog is growing every day.

it looks like it was a software feature in my trickle program to limit the # of trickles to 1000 per minute (which should be plenty). But had things falling behind as there\'s a hacker sending like a million trickles a day. So with this guy clogging up the database, things were falling behind. I hope to have it sorted out shortly.

there\'s a hacker sending like a million trickles a day

I hope I have not been contributing to this overload! I noticed my machine has been trying to send up a trickle every hour or so. If 50 thousand other machines are doing the same thing, could this effect be causing the apparent DOS attack?
However, I turned off overnight and on switch-on this morning was able to upload the pending three or four trickles with no bother - looks like the problem (thanks to all involved) is finally going away.

---