climateprediction.net home page
Slow progress on WU hadsmfub

Slow progress on WU hadsmfub

Message boards : Number crunching : Slow progress on WU hadsmfub
Message board moderation

To post messages, you must log in.

AuthorMessage
old_user559498

Send message
Joined: 5 Mar 09
Posts: 5
Credit: 5,034,381
RAC: 0
Message 38179 - Posted: 25 Oct 2009, 21:46:33 UTC

One of my rigs has been working on a hadsmfub_k1jx_006414807_4 for what seems like an eternity. Does anyone know the average time it takes to complete one of these WUs.
The CPU is an Intel Celron 1200 clocked at 3.2GHz not too much slower than an E6600 @ 3.2GHz on most WUs.
Just wondering if the WU is OK and it will just take a long time to complete or if there is something wrong with it.
sculptor.
ID: 38179 · Report as offensive     Reply Quote
Les Bayliss
Volunteer moderator

Send message
Joined: 5 Sep 04
Posts: 7629
Credit: 24,240,330
RAC: 0
Message 38180 - Posted: 25 Oct 2009, 22:04:14 UTC

Celerons are a budget processor with not much power.
On my Q6600 at stock speed of 2.4GHz, a slab model takes about 2 weeks.
Seriously overclocking processors and expecting a climate model to be unaffected, mostly leads to faulty data.

There are 2 problems with the slab models that can be a problem:
1) If interrupted at the end of a phase, when a LOT of post-processing takes place, (15-30 minutes), they can rewind to the start of the model and start again.

2) They have a tendency to turn into a so called \"ice world\", (because the graphics reverts to the default starting colour of blue), which can be of two types:
A) A \'slow processor\', taking forever to move forward, or
B) A \'fast processor\', where the model speeds up drastically because not all of the data is being created.

If you see this on the graphics image, you can try a restore from a previous (manual) backup, or just abort it.
There\'s a thread about iceworlds here.


Backups: Here
ID: 38180 · Report as offensive     Reply Quote
Profile geophi
Volunteer moderator

Send message
Joined: 7 Aug 04
Posts: 2169
Credit: 64,555,907
RAC: 5,858
Message 38181 - Posted: 25 Oct 2009, 22:38:08 UTC

It looks like your model has indeed turned into a slow-processing iceworld. Les has linked to the thread that describes that. I would abort and get a new model.
ID: 38181 · Report as offensive     Reply Quote

Message boards : Number crunching : Slow progress on WU hadsmfub

©2024 climateprediction.net