I have a sulphur model where the results from phases 2 and 3 seem identical. See http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=1208709

From what I understand the first 3 phases in sulphur should have about the same reaction as a normal slab, with increased CO2 in phase 3 usually leading to increased temperature. Is that correct? Is this then a faulty work unit? Should I abort the model?

---

From what I understand the first 3 phases in sulphur should have about the same reaction as a normal slab, with increased CO2 in phase 3 usually leading to increased temperature. Is that correct? Is this then a faulty work unit? Should I abort the model?

Depending of initial WU\'s parameters, the computed values for temperatures & precipitations may very much or less. Take a look at this graph. If analysis of the data gives a slight increase of 2 to 3 degrees Celsius when CO2 increase, some returned results get a lot warmer than the \'expected\' average, whilst others cool.

I don\'t think you have a faulty WU and your WU is probably more interesting to CPDN scientists than a \'normal & more than usual\' WU. So, please don\'t abort your model.

---

I have a sulphur model where the results from phases 2 and 3 seem identical. See http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=1208709

From what I understand the first 3 phases in sulphur should have about the same reaction as a normal slab, with increased CO2 in phase 3 usually leading to increased temperature. Is that correct? Is this then a faulty work unit? Should I abort the model?

It looks to me this is one of a faulty batch!

May I suggest to backup the model and start a new one, till it is known which batch was faulty.

---

Faulty batch. See this post (and some others in that thread) for more explanation. DaveF said they may be useful anyway for computer comparisons, but you certainly can abort it and get a new one if you so desire.

I had one where all of a sudden the s/TS was super low ... I just let it run and eventually it died.

One of the \"fun\" things in CPDN, and Rosetta@Home is that there is almost no such thing as a \"failure\". The models show something even when they go \"bad\".

The fact that we get credited for the effort doesn\'t hurt ... after all, it is all about the credits! :)

---

*ugh*

I\'d be happy to even reach Phase 2.
All of my Sulphur Models end with Error -161 no later than Trickle 24, no matter what I do or what System they run on :/

By now that are like... 20 already ?
My Results page looks pretty ugly by now, all listed as \"Computing Error\" (the Systems are 100% okay, validating perfect on all other Projects)

---

Scientific Network : 44800 MHz - 77824 MB - 1970 GB

Faulty batch. See this post (and some others in that thread) for more explanation. DaveF said they may be useful anyway for computer comparisons, but you certainly can abort it and get a new one if you so desire.

Thank you.
The thread refered to is a bit unclear on the on the topic of whether to abort the run or not...

Well the runs are still useful (even if there is a measure of redundancy in them), so we don\'t want to insist people kill those runs, but if you want to make best use of your computer\'s processing time, you could kill the run and get a new experiment from Tolu\'s list (which he\'s generating right now). Maybe give it a couple of hours till the jobs are in place.

This quote from DaveF indicates that it would be more efficient to kill the model, but is this still true after 3 phases have already been processed?

Then there is also the issue of when the coupled model will be given a go on the main CPDM site. If that is soon, I may just as well wait for that rather than starting a new sulphur, and then finish this one in the mean time. Does anyone have info on that one?

---

> ... when the coupled model will be given a go ...
> Does anyone have info on that one?

It\'s a secret.
Some time next month. Perhaps.
The only way to find out is to wait until it happens.

And those that can\'t get sulphur to run may have no better luck with coupled ocean.

By now that are like... 20 already ?
My Results page looks pretty ugly by now, all listed as \"Computing Error\" (the Systems are 100% okay, validating perfect on all other Projects)

FalconFly,

The WU\'s that bomb out after the 24th trickle were created between Dec 9th and the 22nd. Everyone of those erorred out after the first phase for everyone.

Additionally, and unfortunately, since you have so many Linux systems, you are experiencing the sulphur 4.23 problems like described in this thread.

So, there is likely nothing wrong with your systems as you surmised. If I were you, I would suspend cpdn on all your Linux systems that might be running 4.23 and wait until a new version comes out.

This quote from DaveF indicates that it would be more efficient to kill the model, but is this still true after 3 phases have already been processed?

If Ph3 is a exact copy of Ph2, than Ph5 will be a copy as well, so running Phase 5 will be a useless waste.

It might be that Phase 4 is someway usefull, but don\'t say it will be!

---

FalconFly,

The WU\'s that bomb out after the 24th trickle were created between Dec 9th and the 22nd. Everyone of those erorred out after the first phase for everyone.

Additionally, and unfortunately, since you have so many Linux systems, you are experiencing the sulphur 4.23 problems like described in this thread.

So, there is likely nothing wrong with your systems as you surmised. If I were you, I would suspend cpdn on all your Linux systems that might be running 4.23 and wait until a new version comes out.

Thanks, so it looks like there\'s not much to do.
I\'ll go ahead and suspend new work on the machines already on sulphur then until the Problem is fixed...

---

Scientific Network : 44800 MHz - 77824 MB - 1970 GB