Another question for Les Bayliss and Saenger, and/or others:
I intend to investigate the climate in Malaga in middle April, about the same time when my run of hadsm3 4.04 will complete the last Phase.
I don't mind leaving the cpu running 24-7, but there will be no internet connection, thus no reporting and other interaction with BOINC head office.
Will this simply resume and catch up when I return and reconnect?

---

There's no problem with leaving networking disabled John.

The only thing you'll need to be aware of when you reconnect to the network is <a href="http://www.climateprediction.net/board/viewtopic.php?t=2811#25308">here</a> on the phpBB forum.

To make sure your system doesn't run out of work when its current model finishes you might also consider increasing the 'Connect to network' setting in your general preferences and force another model to be downloaded by doing a project update ;)

---

"The ultimate test of a moral society is the kind of world that it leaves to its children." - Dietrich Bonhoeffer

&gt; There's no problem with leaving networking disabled John.
&gt;
&gt; The only thing you'll need to be aware of when you reconnect to the network is
&gt; <a> href="http://www.climateprediction.net/board/viewtopic.php?t=2811#25308"&gt;here</a>
&gt; on the phpBB forum.
&gt;
&gt; To make sure your system doesn't run out of work when its current model
&gt; finishes you might also consider increasing the 'Connect to network' setting
&gt; in your general preferences and force another model to be downloaded by doing
&gt; a project update ;)
&gt;

Thyme Lawn: Thanks again. Further to my previous reply; I increased the network connection frequency to 0.5 days, and updated. Then I looked at Les Bayliss and your own pages. Les manages several (4 or more) per day, but you send an astonishing ten per minute. I see that these represent Phases 1 and 2 of probably different programs. Do the trickles interfere with crunching, as someone (Les?) suggested? Is that a Cray you're running?

---

&gt;
&gt; Thyme Lawn: Thanks again. Further to my previous reply; I increased the
&gt; network connection frequency to 0.5 days, and updated. Then I looked at Les
&gt; Bayliss and your own pages. Les manages several (4 or more) per day, but you
&gt; send an astonishing ten per minute. I see that these represent Phases 1 and 2
&gt; of probably different programs. Do the trickles interfere with crunching, as
&gt; someone (Les?) suggested? Is that a Cray you're running?
&gt;
Ten trickles a minute! Wow! what performance!

If we could see the last 20 or 30 it might look even better but if we saw the last 75 then I presume we would see a gap of a dozen hours or something. 15 machines and perhaps 2 to 4 trickles from each. Did you use something like BOINCview to tell them all to update at once?

Even understanding this, it is impressive. ;)

&gt; Thyme Lawn: Thanks again. Further to my previous reply; I increased the
&gt; network connection frequency to 0.5 days, and updated.

I think Thyme meant: If you get a model before you go away, then no problem. If you don't, increase that setting again to something like 14 days to ensure you do get a model before going.

&gt; &gt; Thyme Lawn: Thanks again. Further to my previous reply; I increased the
&gt; &gt; network connection frequency to 0.5 days, and updated.
&gt;
&gt; I think Thyme meant: If you get a model before you go away, then no problem.
&gt; If you don't, increase that setting again to something like 14 days to ensure
&gt; you do get a model before going.
&gt;
OK. I see: so I am misinterpreting connection frequency. I will try that at your suggested 14 days.
Thanks Crandles.

---

&gt; &gt; &gt; Thyme Lawn: Thanks again. Further to my previous reply; I increased
&gt; the
&gt; &gt; &gt; network connection frequency to 0.5 days, and updated.
&gt; &gt;
&gt; &gt; I think Thyme meant: If you get a model before you go away, then no
&gt; problem.
&gt; &gt; If you don't, increase that setting again to something like 14 days to
&gt; ensure
&gt; &gt; you do get a model before going.
&gt; &gt;
&gt; OK. I see: so I am misinterpreting connection frequency. I will try that at
&gt; your suggested 14 days.
&gt; Thanks Crandles.

Follow-up : for Crandles and Thyme Lawn:
You are obviously right on the money. It was updated and immediately threw more work at us. Am I getting this right?

&gt;

---

&gt; &gt; &gt; &gt; Thyme Lawn: Thanks again. Further to my previous reply; I
&gt; increased
&gt; &gt; the
&gt; &gt; &gt; &gt; network connection frequency to 0.5 days, and updated.
&gt; &gt; &gt;
&gt; &gt; &gt; I think Thyme meant: If you get a model before you go away, then no
&gt; &gt; problem.
&gt; &gt; &gt; If you don't, increase that setting again to something like 14 days
&gt; to
&gt; &gt; ensure
&gt; &gt; &gt; you do get a model before going.
&gt; &gt; &gt;
&gt; &gt; OK. I see: so I am misinterpreting connection frequency. I will try that
&gt; at
&gt; &gt; your suggested 14 days.
&gt; &gt; Thanks Crandles.
&gt;
&gt; Follow-up : for Crandles and Thyme Lawn:
&gt; You are obviously right on the money. It was updated and immediately threw
&gt; more work at us. Am I getting this right?

HERE IS THE INFO SENT ON "MESSAGES" AFTER THE UPDATE TO 14 DAYS. IF YOU HADN'T CAUGHT MY ERROR ALL MAY HAVE BEEN FOR NAUGHT. I HOPE THIS IS OF ASSISTANCE TO OTHERS.
(hadsm3 4.10 is sitting there waiting for the completion of 4.04 ??)

climateprediction.net - 2005-03-25 09:39:13 - Sending request to scheduler: http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
climateprediction.net - 2005-03-25 09:39:16 - Scheduler RPC to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi succeeded
climateprediction.net - 2005-03-25 09:39:16 - General preferences have been updated
--- - 2005-03-25 09:39:16 - General prefs: from climateprediction.net (last modified 2005-03-25 09:29:22)
--- - 2005-03-25 09:39:16 - General prefs: no separate prefs for home; using your defaults
climateprediction.net - 2005-03-25 16:37:50 - Sending request to scheduler: http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
climateprediction.net - 2005-03-25 16:37:53 - Scheduler RPC to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi succeeded
climateprediction.net - 2005-03-25 16:37:53 - General preferences have been updated
--- - 2005-03-25 16:37:53 - General prefs: from climateprediction.net (last modified 2005-03-25 16:34:29)
--- - 2005-03-25 16:37:53 - General prefs: no separate prefs for home; using your defaults
--- - 2005-03-25 16:37:53 - May run out of work in 14.00 days; requesting more
climateprediction.net - 2005-03-25 16:37:53 - Requesting 1994364 seconds of work
climateprediction.net - 2005-03-25 16:37:53 - Sending request to scheduler: http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
climateprediction.net - 2005-03-25 16:37:57 - Scheduler RPC to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi succeeded
climateprediction.net - 2005-03-25 16:37:57 - Started download of hadsm3_4.10_windows_intelx86.exe
climateprediction.net - 2005-03-25 16:37:57 - Started download of hadsm3data_4.10_windows_intelx86.zip
climateprediction.net - 2005-03-25 16:38:39 - Finished download of hadsm3_4.10_windows_intelx86.exe
climateprediction.net - 2005-03-25 16:38:39 - Throughput 48948 bytes/sec
climateprediction.net - 2005-03-25 16:38:39 - Started download of hadsm3um_4.10_windows_intelx86.zip
climateprediction.net - 2005-03-25 16:39:13 - Finished download of hadsm3data_4.10_windows_intelx86.zip
climateprediction.net - 2005-03-25 16:39:13 - Throughput 50295 bytes/sec
climateprediction.net - 2005-03-25 16:39:13 - Started download of hadsm3se_4.10_windows_intelx86.zip
climateprediction.net - 2005-03-25 16:39:29 - Finished download of hadsm3um_4.10_windows_intelx86.zip
climateprediction.net - 2005-03-25 16:39:29 - Throughput 48528 bytes/sec
climateprediction.net - 2005-03-25 16:39:29 - Finished download of hadsm3se_4.10_windows_intelx86.zip
climateprediction.net - 2005-03-25 16:39:29 - Throughput 48522 bytes/sec
climateprediction.net - 2005-03-25 16:39:29 - Started download of globe.tga
climateprediction.net - 2005-03-25 16:39:29 - Started download of globe.rgb
climateprediction.net - 2005-03-25 16:40:03 - Finished download of globe.rgb
climateprediction.net - 2005-03-25 16:40:03 - Throughput 45447 bytes/sec
climateprediction.net - 2005-03-25 16:40:03 - Started download of user_logo
climateprediction.net - 2005-03-25 16:40:06 - Finished download of user_logo
climateprediction.net - 2005-03-25 16:40:06 - Throughput 13290 bytes/sec
climateprediction.net - 2005-03-25 16:40:06 - Started download of cpdn_logo
climateprediction.net - 2005-03-25 16:40:09 - Finished download of cpdn_logo
climateprediction.net - 2005-03-25 16:40:09 - Throughput 13247 bytes/sec
climateprediction.net - 2005-03-25 16:40:09 - Started download of 14s6_000073405.zip
climateprediction.net - 2005-03-25 16:40:11 - Finished download of 14s6_000073405.zip
climateprediction.net - 2005-03-25 16:40:11 - Throughput 9570 bytes/sec
climateprediction.net - 2005-03-25 16:40:51 - Finished download of globe.tga
climateprediction.net - 2005-03-25 16:40:51 - Throughput 76512 bytes/sec

&gt;
&gt; &gt;
&gt;

---

&gt; &gt; Thanks Crandles.
&gt;
&gt; Follow-up : for Crandles and Thyme Lawn:
&gt; You are obviously right on the money. It was updated and immediately threw
&gt; more work at us. Am I getting this right?
&gt;
No problem. That should keep your computer busy ;)

&gt; Thyme Lawn: Thanks again. Further to my previous reply; I increased the
&gt; network connection frequency to 0.5 days, and updated. Then I looked at Les
&gt; Bayliss and your own pages. Les manages several (4 or more) per day, but you
&gt; send an astonishing ten per minute. I see that these represent Phases 1 and 2
&gt; of probably different programs. Do the trickles interfere with crunching, as
&gt; someone (Les?) suggested? Is that a Cray you're running?

No such luck :(

But I do have 11 hyper-threaded P4s :)

I disable networking overnight and at weekends, which is why you see bursts of trickles.

---

"The ultimate test of a moral society is the kind of world that it leaves to its children." - Dietrich Bonhoeffer