The run-times are probably correct, and for some work units, the CPU times are a little less than the run-times. I assume those are correct. But there are other work units where the CPU times are ridiculously low, such as this one:

Task ID
click for details
Show names Work unit ID
click for details Computer Sent Time reported
or deadline
explain Status Run time
(sec) CPU time
(sec) Claimed credit Granted credit Application

19093944 9672778 1256552 23 Nov 2015 14:13:51 UTC 7 Dec 2015 15:48:22 UTC Completed 1,203,649.89 987.22 9,559.58 9,559.58 UK Met Office HadAM3P and HadRM3P model with MOSES II and TRIFFID Europe v7.01

My guess is that I am running on a 64-bit machine and only the least significant 32-bits are being reported. Cloud this guess be correct?

I am running Red Hat Enterprise Linux 6.7 with the 32-bit compatibility libraries available.

---

The model name for reporting purposes is e.g. hadam3prm3pm2t_eu_pfop_2002_1_009605406_2

Or you can use the Task ID number e.g. 18269335

Neither of these need interpreting or a long search to find the task in question.

As for the "low times", you may be talking about the models where only part of the modelling time is being reported.
The time should be: (the time for the regional model) plus (the time for the rest of the world). But only one of these is being sent back.
This is an old problem, and I've forgotten the details. Plus the posts about it were on the beta site, which is now defunct.

But the data files themselves are all being correctly returned.