CP shows me that i don't have enough disk space but i have almost 1 gb free for Boinc

---

Hallo Peter,
I think 1GB should be OK.
But you have to have the right preferences.
As you are german, look <a href="http://climateapps2.oucs.ox.ac.uk/cpdnboinc/forum_thread.php?id=1782">hier</a> for my setting. (OK, I've got a little more GB, but the gist should be sufficient;-)
<a href="http://setiweb.ssl.berkeley.edu/forum_thread.php?id=11396">Here</a> is a thread at Seti's.
You can as well look in <a href="http://boinc-doc.net/index.php">PD Bucks Manual</a>.

---

Grüße vom Sänger

Patrick

First, welcome to CPDN, we users should be able to help you get things up and running soon.

The <a href="http://climateapps2.oucs.ox.ac.uk/cpdnboinc/quick_faq.php#g">FAQ</a> contains some information on how much space each model needs. It needs about 600Mb for each full model run and about 330Mb of information will be left on your computer once a model run is finished and the partial information uploaded.

That being said the 1 Gb that you have free will allow you to run about 1 complete model before your Hd becomes full. You should look into clearing up some space on your Hd then defragging the Hd to have as much contiguous space possible.

Searching the net with Google should yeild you instructions on clearing up space on your Hd, other than buning CD's of your information.

Have to disagree here Dave... Looked at my model the other day and it was eating around 850-900Mb of HDD space during phase 3.

---

<img src="http://boinc.mundayweb.com/one/stats.php?userID=444&amp;trans=off">

&gt; Have to disagree here Dave... Looked at my model the other day and it was
&gt; eating around 850-900Mb of HDD space during phase 3.

I find that surprising. I have a machine running two models under HT at the moment, one halfway through phase 2 and the other two thirds of the way through phase 3. The total size of the BOINC folder is 748 MB.

Are you sure it didn't include folders from earlier runs, which are archived in the BOINC projects folder by default?

---

&gt; &gt; Have to disagree here Dave... Looked at my model the other day and it
&gt; was
&gt; &gt; eating around 850-900Mb of HDD space during phase 3.
&gt;
&gt; I find that surprising. I have a machine running two models under HT at the
&gt; moment, one halfway through phase 2 and the other two thirds of the way
&gt; through phase 3. The total size of the BOINC folder is 748 MB.
&gt;
&gt; Are you sure it didn't include folders from earlier runs, which are archived
&gt; in the BOINC projects folder by default?

I was just quoting the FAQ that was set up in September. There may have been some changes which increases the Hd requirements, and or if as Andrew stated other information downloaded. Mine has been up to 5 Gb but that was including completed models. Once I archive the complete runs it is generally 1.2 Gb. A WinXP Athlon machine with 1 completed model mid phase 2 is using 713Mb. The about 600Mb figure seems right.

As far as the figure I quoted was concerned, it was just the figure used by CPDN according to the breakdown provided under the disk usage tab on the Windows client. Whether this includes old runs or not, I'm not sure. I guess it would. I probably do have two dead models in there somewhere. Is there a fool-proof way of killing them off for he who is not sure about the inner filesystem of the BOINC client? I don't particularly want to lose my current model. Not with an ETA of 4-5 days.

Maybe I'll leave it until after the current run has completed and do it when I do the 4.13 -&gt; 4.19 update.

---

<img src="http://boinc.mundayweb.com/one/stats.php?userID=444&amp;trans=off">

Gareth,
to check for dead or complete models:
Look in your BOINC directory (normally in program files if you're using Windows)
You'll see a projects folder.
In that folder there is a climateprediction.net subfolder.
Inside that are your models.
The current model shows as two files andd a folder:
one .xml file, one .zip file and a folder with the same name.
I'm not sure about dead models, but a finished model will just have the folder with a lot of zip files in it.
What I do:
When a new model has sent its first trickle and the previous model doesn't show in the work tab in the BOINC interface (that's to make sure) I move the completed model to a backup I keep on another disk.
That frees up space for BOINC (you can see that in the disk tab) and keeps the zipped files in case they're required for further studies.

---

Forum search Site search

Thanks for the advice... Killed off 300Mb odd of files stuck in two dead WUs. Approx used space now down to 650 - 680Mb odd.

---

<img src="http://boinc.mundayweb.com/one/stats.php?userID=444&amp;trans=off">

&gt; The FAQ
&gt; contains some information on how much space each model needs. It needs about
&gt; 600Mb for each full model run and about 330Mb of information will be left on
&gt; your computer once a model run is finished and the partial information
&gt; uploaded.

Every model will leave 330 MB on my hard drive? The FAQ does not mention that AFAIK.
Is that information needed later of can I safely discard it?
(I have enough space for my two CPDN WUs but not enough to keep all the results.)

Friedrich

---

I love CPDN!
--

It isn't essential to keep the files if you really don't have room. But if you can, either on another drive or on CD/DVD, you should. Occasionally scientists ask if you can upload them, and you might develop an interest in the subject and want to compare runs yourself. There is a new version of the advanced visualisation tool expected soon, and you can also use CPView for this.

Part of the attraction of climateprediction is that you can explore the science yourself.

---

[network problems caused an accidental repost that I deleted - F.]

---

I love CPDN!
--

Friedrich,
The info about the 330M of files IS mentioned in the FAQ, but it's 'wearing dark glasses', so it's hard to recognize.
This is the relevant part:

If I go to BOINC, should I uninstall classic?

It is possible to run classic and BOINC in parallel so it is not necessary to uninstall classic.
We have a quasi-archive uploder mechanism for uploading archived experiments.
** If you have a high speed network with high throughput AND COST is not objective,
** you can upload the archived experiments here or here. - ( requires some sort of browser plugin )
** Otherwise if possible please keep the backup of the archived run,
the scientists may want to do analysis on them.
If you simply want to free up disk space, then you can go ahead and delete the experiment
folder within the archive directory.

All mine are on cds, in case of a hd accident.

Les