I run oneWU each 30 min.
So, I only can store for less than 10 hours.
Is it a limitation ? I can undertand uch wazy, but with my CR and RAC, I think i crunch enouh to how I am not a "ghost", or someone who try and the delete without crunching.
Of coure my preference (general and project) are set on 2 days. to not overload server.
I see some host have a lot of WU. What do I wrong ???
Greetings from Belgium

---

The limitation is 20 WU per single GPU and 2 WU per single CPU core.

Hello Sergei,
I undertand the reasons. But perhaps is it posible to send amount WU depending RAC, or average turnaround time ?
On SETI it are 100WU/GPU and 100/CPU.

If I disconnect my host more than 10 hours, I run out work
I repeat and agree the reasons of limitation of 20WU. But .....
I have a modest mid-range GPU and mid-range CPU.
Some of us have very high end GPU. Si they runs WU very fast.
Probably after 2 hours all WU are done ! I project is few hours down , we can not more crunch.
It i why I insist to reconsider understable limitations dependeing RAS and/or avg turnaround time
Best regards and thanks for answer
[/url]

---

The project has never been down for more than a few minutes so far, it's running on a dedicated server in a data center. Is it a real issue for you? Do you often get disconnected for many hours?

"Sergei Chernykh" wrote:

The limitation is 20 WU per single GPU and 2 WU per single CPU core.

In this calculation, the number of CPU cores is the minimum of

- the number of active CPUs set in the boinc-client, and
- the "Max # CPUs" setting in the project preferences on this web site,

right?

No, the number of CPU cores that count for this restriction is what you can see in your Computers belonging to page.

P.S. Actually, BOINC client settings can change this number. But "Max # CPUs" setting shouldn't influence it.

It does though... When I joined I set "Max # CPUs = 8" and got only 16 tasks in progress per host. Now I removed the setting temporarily and received more.