As a result I received the message "Error of calculations"
You that, mocked?

---

Some of the climate parameter sets can become unstable and lead to crashes, but there are a number of things you could check:

• Is your BOINC data directory excluded from anti-virus scans? Some AV programs put an exclusive lock on files as they are scanned. If that coincides with the application attempting to write to the file it can cause a crash.
  
  
• Frequent rapid stops and starts can also cause problems. If this is happening you could change your global preferences to keep applications in memory (they run at low priority and will be swapped to disk if your computer needs the memory for something else. If you want you could also change the "Suspend work if CPU usage is above" setting to 0%. That would allow BOINC to continue using all of the idle time when other programs are using more CPU time than the current threshold.
  

The time spent on that task wasn't totally wasted; the trickle list shows the task had used 94 hours of CPU time when it returned its 8th trickle. Each of those trickles will have received credit.

---

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

It is also worth suspending the work unit and exiting boinc using the file > exit menu item. This seems to affect the full resolution ocean models more than the regional ones in my experience or it may just be that because they are much longer they get more chances of us crashing them? But rest assured, the data that goes back still gets used even if the model does not finish.

Dave