cumulative RAM usage increase

[uyaem]

i don't understand what sensors tab is.

From https://www.techpowerup.com/gpuz/ you can download GPU-Z, a tool with all kinds of useful information on the GPUs in your system; temperatures, fan speeds, you name it - and that is located on the Sensors tab.

i am thinking too that this may be related to me having 2 gpus running. do you think many donors are running multiple gpus?

I don't think it's all that common, but there are enough power users with anywhere between 2 and 6.

[PantherX]

...i am thinking too that this may be related to me having 2 gpus running. do you think many donors are running multiple gpus?

Please note that while uyaem is correct, the context is that most Donors tend to stick with multiple AMD or Nvidia GPUs in a single system.

In your case, you have an AMD and a Nvidia GPU in a single one:

Code: Select all

19:43:00:             OS: Windows 10 Enterprise
19:43:00:        OS Arch: AMD64
19:43:00:           GPUs: 2
19:43:00:          GPU 0: Bus:1 Slot:0 Func:0 NVIDIA:5 GM204 [GeForce GTX 970] 3494
19:43:00:          GPU 1: Bus:6 Slot:0 Func:0 AMD:5 Ellesmere XT [Radeon RX
19:43:00:                 470/480/570/580/590]
19:43:00:  CUDA Device 0: Platform:0 Device:0 Bus:1 Slot:0 Compute:5.2 Driver:11.0
19:43:00:OpenCL Device 0: Platform:0 Device:0 Bus:6 Slot:0 Compute:1.2 Driver:3004.8
19:43:00:OpenCL Device 1: Platform:1 Device:0 Bus:1 Slot:0 Compute:1.2 Driver:445.87

Over time, V7 got a lot better with handling multiple vendor GPUs in a single system but it's possible that there are some undiscovered bugs. Your case of having multiple vendor GPUs in a single system is rather rare.

[cayenne187]

thx uyaem gpuz is awesome!

[cayenne187]

current wu are both heavy users.
14253 (1881, 1, 25) is accumulating ram use after uploading every ~15 minutes.
14562 (0, 312, 48) just started and is doing the same.
these are the type wu that do it. stops folding every 15 minutes for 30 seconds, big spike in cpu and ram and after gains 50-100mb ram.
Happy Memorial Day everyone! Stay safe!

[bruce]

P14253 is the biggest protein that I've ever seen. Yes, it's bound to be very demanding. P14562 isn't quite as big, but it's close. My wish for you is to never get those two WUs at the same time again.

I suspect that the every ~15 minutes processing is normal (although the frequency probably changes depending on the protein). That sounds like the part of the processing when the FAHCore writes a checkpoint to disk. (from which it can resume if it crashes or is paused). It does some extra validation steps, too, to verify that the checkpoint is coherent.

[cayenne187]

thx bruce. i have no worries. i added 8gb ram. i am just providing info to sort out bugs and make system work better. when i had 8gb ram that was not enough. running stable now again. also throttled back gtx 970 a bit. it uses 150 watts vs. th rx570 using 80. running nice and cool at 350 watts at the plug and 5-600,000 points a day. i'm good with that. nice to be busy all the time!

[bruce]

This system's M/B only takes 6 GiB and I'm running 3 slots.