I'm sure it must be especially frustrating to you when completed tasks are queued up on volunteers' computers because the server isn't in a position to accept the uploads and/or reports, and then again because the BOINC clients have gone into an extended backoff because of the connection failures.

I've just manually retried and successfully reported two tasks which had completed overnight, and got replacement tasks on both machines: at least they started running almost immediately, so not too much time was lost. But do your IT guys have any idea why the new server seems to become unresponsive so much more often than the old one?

... But do your IT guys have any idea why the new server seems to become unresponsive so much more often than the old one?

I hope we don't have to get used to server outages happening every other week.
This time we were lucky anyway as it happened on a Thursday evening, so due to the fact that Friday is a workday, the problem was fixed this late morning (while last time the server was down from Friday evening till Monday noon).

Well, even my PCIe2x16 conncected gtx1080 runs at pretty 95% per GPU-Z... which is a very pleasing figure. The BNBS accommodate the fast Pascal I presume and reduces the WDDM influence. Anyway, I am happy with the throughput.

Exactly. There's unusually little interaction between the CPU and the GPU by these workunits.

For comparison, I'm seeing 40% PCIe usage on my GTX1060-3GB running a PABLO_adaptive task (with the GDDR@8GHz & GPU at ~1950MHz on Linux; ~95% GPU Utilization. PCIe2 x16, while only running 1 CPU task on a tri-core AMD).

Well it takes him 11 days to complete a BNBS2, now it should take slightly over 24 hours, so over 10 times as fast. I like to exaggerate as well :)

If it takes over 5days there really isn't much point 'contributing'.

I'm sure it must be especially frustrating to you when completed tasks are queued up on volunteers' computers because the server isn't in a position to accept the uploads and/or reports, and then again because the BOINC clients have gone into an extended backoff because of the connection failures.

I've just manually retried and successfully reported two tasks which had completed overnight, and got replacement tasks on both machines: at least they started running almost immediately, so not too much time was lost.

The continuation of this problem calls into question the credit/bonus system, unless of course you don't value it much/at all. When crunchers complete work quickly but can't upload & report they miss the bonus because the server isn't available, & BM backs off the retries. It's not the fault of the cruncher, doesn't help the project get through work, or retain crunchers! GPUGrid Stats regarding the fastest GPU's/systems/contributors become fickle, further frustrating some crunchers.

But don't worry, there's only 2.5min to go!

---

FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

My only disappointment is that my GPUs are showing their age: I cannot use the two GTX 650 Ti beasts any more and the GTX 660 Ti GPUs have very long run times. I can't afford to buy faster GPUs.....

---

John

You could sell them

Also, I just posted 3 times because the server was too slow and took my post reply command 3 times. This would normally not be a problem but there is no way for us normal users to delete our own messages.