A few people have asked about why there are no more slab models, so herewith a short story.

Slab was to find stable values for the various parameters, for use in the next phase, \'experiment 2\', (as mentioned in the basic Climate Science page to the left of here), which is also known as the Couple Ocean experiment. (\'slab\' refers to the fact that the ocean is a fixed object, and doesn\'t interact with the atmosphere.)
The scientists / researchers now have over 100,000 of these slab models.
One of the other needed variables are sulphate particles, and how they interact.
So two extra phases have been added to the standard slab model, in a similar manner to CO2: current levels of sulphates, and double this level.
It seems that the researchers have decided that the data they have received from the early models returned is interesting, and have asked for lots more of these models. So, to \'force\' the processing of sulphur, the slab models have been \'turned off\'.
Slab is still being processed by people, who, for one reason or another, can\'t run BOINC. And also by people doing the Open University\'s Climate course.

In an \'alpha/beta area\', a few dozen people are running another part of the lead up to \'coupled ocean\'; this is called \'spinup\', and is a single phase, 200 year run. It trickles once per model month, (on my P4 3.2GHz machine running at 3.4GHz, every 68 minutes), and will take approx. 4 months to complete. And, from when I started, I have 2 months to do it. :(

So, sometime in February, the coupled ocean part of the project is expected to start. But I think that sulphur will continue for some time, to build up a large database of results.

And, just to depress people even more, the coupling between the ocean and atmosphere parts has proven very tricky to get stable, so if your computer isn\'t stable enough to run sulphur, you stand little chance of getting coupled to run.

So now is a good time to get computers fine tuned, and problems sorted out.

Thanks for the information !

---

There are also those of us that have not finished running all of our slab models (I have 2 to go; one with 1.5 and the other with 3.6 days to go - of course, on the calander that will be longer), but am getting close to completing my first Sulfur cycle models; with one having only 10 days to go.

Thanks for the update Les.

The longer run times are a little depressing, especially if instability has increased. If the slab models are still of use, you may want to consider a screening of compute speed so that some of the slower systems can still participate. I mean, I went from less than 30 to nearly 90 days on my models. I don\'t mind as my systems are relatively high end (though there are faster ones out there). I guess I just hate to think of those that would leave the project because they cannot do the longer models.

Longer term this would also mean that you may have multiple tiers ...

---

Carl and Tolu tried screening right at the start of the public release of sulphur, but it didn\'t work. Some sort of problem with the server software I think, and they didn\'t have time to experiment.

Bound to be a lot of slabs still in the pipeline, but sooner or later the project has to move on.
But \'slower\' systems would still work IF people made them stable. It\'s just so hard to help people do this, with minimum information, and an unwillingness / inability by users to upgrade parts of their hardware / software to do it.
Part of this problem is that users are running on work computers, and computers belonging to friends and relatives.

Part of this problem is that users are running on work computers, and computers belonging to friends and relatives.

Oh, the temptation ...

Shame I can\'t get any help like this ... sigh ... I guess I will have to stick with buying my own systems still ...

---

Hey Les. Is it known how much disk space the next experiment will require?

> Hey Les. Is it known how much disk space the next experiment will require?

Not by me. I\'ve posted just about everything I\'ve come across, except for the BBC project, due to launch in January. Which is apparently something to do with putting very stable software on cd, for distribution to the public.

However, I was looking at the Climate Science pages, and in one place it says:

\"We will start a bunch of experiments in 1950 and force them with historical data for fifty years.\"

As the spinup is one long run of 200 years, with 52K trickles every model month, it occured to me that perhaps this is what is inferred in the above quote, but for 50 years.
Just thought: I\'ve just gone past 50 years in spinup, so have a look at hd. Not much space used; 165Megs, according to ZtreeWin. And Carl says on the beta site that a few hundred Megs is enough.

But you need to send about 1meg of data per day. I guess this will change when it goes live, with people switching between lots of projects. But it is heavy on frequent trickles, so for people with dialup, these will accumulate until the next contact.
So, slower computers may be back in business.
All of which is speculation.

I think the main problem will be the same as now: you MUST have a stable computer.

Not by me. I\'ve posted just about everything I\'ve come across, except for the BBC project, due to launch in January. Which is apparently something to do with putting very stable software on cd, for distribution to the public.

I was hoping that Boinc was going to be used in the next stages as it is now becoming stable with 5.2.13 (4.xxx was *not* great) Plus you can update us to your newest version with out another CD being released. (Unless it is going to be an iso ready for download (-: )

---

> Hey Les. Is it known how much disk space the next experiment will require?

Not by me. I\'ve posted just about everything I\'ve come across, except for the BBC project, due to launch in January. Which is apparently something to do with putting very stable software on cd, for distribution to the public.

However, I was looking at the Climate Science pages, and in one place it says:

\"We will start a bunch of experiments in 1950 and force them with historical data for fifty years.\"

As the spinup is one long run of 200 years, with 52K trickles every model month, it occured to me that perhaps this is what is inferred in the above quote, but for 50 years.
Just thought: I\'ve just gone past 50 years in spinup, so have a look at hd. Not much space used; 165Megs, according to ZtreeWin. And Carl says on the beta site that a few hundred Megs is enough.

But you need to send about 1meg of data per day. I guess this will change when it goes live, with people switching between lots of projects. But it is heavy on frequent trickles, so for people with dialup, these will accumulate until the next contact.
So, slower computers may be back in business.
All of which is speculation.

I think the main problem will be the same as now: you MUST have a stable computer.

Not sure whether this will just add more confusion but anyway:

The current coupled spinup doesn\'t use much hard disk space. However I don\'t think they need a detailed history and lots of files are deleted (this has caused a problem or two in getting the code stable). They will need a detailed history for the 50 year hindcast. Sulphur is 5*15 years =75 years. Hindcast is only 50 years but will also have to store data for the ocean. So the hard disk space will definitely be at least 2/3rds of sulphur hard disk space. I doubt it would be more than double the hard disk space and my assessment is more likely than not to be less than 50% more than sulphur.

Unfortunately there is some possibly bad news on the length of runs:

This stems from the point that the next step isn\'t the hindcast, there is a transient spinup to do between the spinup and the hindcast. If the detailed history is not required, like for the spinup, then the HD requirement shouldn\'t be too onerous 300MB ish perhaps.

My knowledge of the transient spinup is very incomplete and dodgy. I am unsure if the 4.11 spinup runs are alpha/beta tests of the transient spinups. If so, I think this implies the transient spinups are also 200 model years and they will be very long like the current spinups.

If 4.11 spinup runs are not alpha/beta tests of transient runs then I have no idea of length. I think the late Jan/Feb launch is a public release of the transient spinup. However I am unsure, so please don\'t critisize me too heavily if I have got this wrong. (Also hope I am allowed to be saying these things.)

If Les and I are confused/disagree then sorry about this. I did say this post might just add confusion.

---

Visit BOINC WIKI for help

And join BOINC Synergy for all the news in one place.

I was hoping that Boinc was going to be used in the next stages as it is now becoming stable with 5.2.13 (4.xxx was *not* great) Plus you can update us to your newest version with out another CD being released. (Unless it is going to be an iso ready for download (-: )

I am sure it is going to be BOINC.

---

Visit BOINC WIKI for help

And join BOINC Synergy for all the news in one place.

Yes, I\'m sure about BOINC too.
The BBC project is TOTALY separate, and nothing to do with people already running here. But it IS taking up Tolu\'s / Carl\'s time, which is why I mentioned it. (It has also been mentioned on the php forum.) Watch the BBC in late January for the official launch.

Only a couple of month\'s to go, and then we\'ll ALL know the situation.
In the mean time, it\'s just sulphur, unless people want to go non-BOINC, and run classic so that they can crunch slab.

just some more \"preview\" of the transient coupled model:

1) we will have it out in beta test in about two weeks, and it will be available from the \"regular\" (i.e. this) CPDN site as well as a separate site BBC viewers will come in from (but they will be a separate \"pool\" of users).

2) it\'s quicker than the slab & sulphur as far as timesteps per second, but as Les mentioned it\'s 160 years (+ 6 days, for 4147632 total timesteps). It will be from the years 1920 to 2080 -- so will span 80 years \"historical\" (hindcast) and 80 years in the future (forecast) -- which they tell me is a common range for studies, IPCC reports etc.

3) we were going to split it up into two 50-year jobs, then two 80-year as we figured that would best match the literature. But there\'s a lot of trouble with say, somebody doing workunit 0j50_0004322 on a Linux 64-bit box, and then for the final 80 years a Win 32-bit box gets it, so it\'s not a well \"controlled\" thing. So having the \"full monty\" 160-year seemed the best

4) it runs about as fast as a spinup, so say the fastest machines, i.e. P-IV 3.8GHz will do the full run in 2.5 months.

5) I didn\'t think it was more unstable than the others, but maybe it is!

6) I am moving from the old \"CPDN hogs your whole hard disk\" paradigm. In the old days we fancied everyone running CPDN would do perhaps 1 model, and it would be nice to keep your data around locally for advanced visualization etc. But I don\'t think that fits in well with BOINC, as we ended up using too much space, people end up doing 10 models a year etc. So I will have the model \"clean up after itself!\" The trickles will provide plenty of information at monthly global & regional scales.

7) the trickles will be smaller than the spinups but more than the slab runs, probably about 100KB per day (since a meg per day is too much for the general public!)

8) there will be 16 uploads (intermediate uploads every 10 years) -- so even if you crash in the middle or towards the end we will have a lot of valuable information. Each upload will be about 3-5MB Max (so cumulatively it will be a lot, i.e. 50-80MB). Which means I hope the BBC or someone buys us more upload servers! ;-)

Hi Guys

It was mentioned in the thread that slab models are being replaced with Sulpher and Exp 2 but many models are still in the pipeline.

My system currently has three slab in the queue, (don\'t know why it got so many models) and I am working my way through them. Should I abort these models in favor of Sulpher and Exp 2.

My system has completed 1 Sulpher already.

Suggestions appreciated
Cheers
Peter

---

Don\'t abort! As long as they are running OK, keep going.
The same with sulphur. Exp 2 won\'t be starting for a while, and sulphur models will be coming in for a year from some people.
I hope to eventually finish my sulphur, suspended at trickle 100 so that I could run spinup. :(

yeah, I wouldn\'t abandon the workunits, at the most (i.e. if you\'re just dying to run a sulphur cycle or spinup or coupled model in a few weeks) suspend and finish later. We\'re becoming more like other projects where we\'ll come up with workunits (for the transient CM3/coupled model coming out soon) where we will regenerate workunits based on past results -- so the slab & sulphur results will eventually appear as \"start dumps\" and/or \"flux adjustments\" for the big future forecast runs (the 1920 to 2080 hindcast/forecast runs that will be the \"grand finale\" of CPDN over the next year or two).

so the slab & sulphur results will eventually appear as \"start dumps\" and/or \"flux adjustments\" for the big future forecast runs (the 1920 to 2080 hindcast/forecast runs that will be the \"grand finale\" of CPDN over the next year or two).

Ok, now I find this depressing... does this mean I can only look forward to at most 2 more years of CPDN?

Oh, and on that disk clean-up, make it an option... I still want to be able to look at my models and I don\'t mind the disk space. Especially in light of the new \"bombshell\" you just laid on me ...

Guys, I suffer enough from my day-to-day struggles with depression ... don\'t add to it! :)

---

...the big future forecast runs (the 1920 to 2080 hindcast/forecast runs that will be the \"grand finale\" of CPDN over the next year or two).

Ok, now I find this depressing... does this mean I can only look forward to at most 2 more years of CPDN?

Time goes different in Oxford so it\'s not 2 years of Earth time :-)
Atmospheric models, High resolution models, regional forecast...there is still plenty to go.
And, as Carl sayd, \"we will regenerate workunits based on past results\". There will be re-runs, re-checkings.

Carl, thanks for the update and keeping us informed. Many things got clear, some are still foggy to me (separated \'pool\', web etc.).

---

<i>phpBB forum for CPDN, all are </i><a href="http://www.climateprediction.net/board">invited</a>

Yes thanks for the updates Carl.

Also very sorry about my post which was billed as adding confusion but was actually just full of misinformation.

---

Visit BOINC WIKI for help

And join BOINC Synergy for all the news in one place.

just some more \"preview\" of the transient coupled model:

1) we will have it out in beta test in about two weeks, and it will be available from the \"regular\" (i.e. this) CPDN site as well as a separate site BBC viewers will come in from (but they will be a separate \"pool\" of users).

Is there a new date for the start of the coupled beta?

---

Linux Users Everywhere @ BOINC

Yes.
Now running.