LTT Folding Team's Emergency Response to Covid-19

[bmichaels556]

1 hour ago, Gorgon said:

Hey I resent that.

 

I don't have a problem.

 

I can stop anytime I want.

 

I've even joined a 12-step program.

 

I mean I did stop buying 2070 Supers after 5 of them and I'm trying to change my ways.

 

See, I just bought 2 5700XTs today

 

Just one more quick GPU for the road.

 

No, I'm OK, I can still drive ...

Holy hell, I need some more money for folding lmao. Teach me your ways, Sensei!  

[Shigeru]

4 minutes ago, bmichaels556 said:

Holy hell, I need some more money for folding lmao. Teach me your ways, Sensei!  

Or rather how to make a bit of money actually... with everything lockdown in Argentina since last week all the jobs I was couting on ended cancelled... but trying to not think too much about economics right now...

[Tad Bittoomuch]

35 minutes ago, TheAmazingDave said:

What platform? I've been trying to get GPU passthrough working on my ESXi host for a few days now with zero luck.

I might be able to save you some headache.

To pass through a GPU with ESXi you need:

 

a) an Nvidia enterprise licence and
b) professional GPU (ESXi doesn't have drivers for consumer GTX/RTX cards)

 

Also as far as I'm aware, it won't work at all with AMD gpus.

 

Source: I asked a a friend who works with VMware/ESXi every day and knows about this stuff
 

[son_of_ottie]

2 minutes ago, orion3814 said:

i sure wish there were more work units right now,  my heaters are hardly flexing..

Not flexin' but I have done 10 WUs today for almost a million points. I'm using GPU only and I fold 24 hours a day uninterrupted. I have a factory overclocked Zotac Gaming GeForce RTX 2070 AMP Extreme Core GPU but don't let all the extra words fool ya! It's still not a Super.
BUT yesterday I had basically nothin'!
What I DID do was follow instructions on another thread here and add a "beta" flag! Still, I'm not sure if THAT is what's keeping the WUs coming. Because that change didn't do anything right away.
What DID get my GPU hummin' again was stopping and re-starting the folding on the web controller. So give that a shot. Reboot before you do if you wanna.
Then have a drink like ME!
 

[son_of_ottie]

7 minutes ago, Shigeru said:

Or rather how to make a bit of money actually... with everything lockdown in Argentina since last week all the jobs I was couting on ended cancelled... but trying to not think too much about economics right now...

Almost the whole working world is in that same boat, my friend. So not worrying is THE WAY to go. Use this time to do other stuff you wanna do...even if it's re-reading Harry Potter!

[pbx2]

9 minutes ago, Tad Bittoomuch said:

a) an Nvidia enterprise licence and
b) professional GPU (ESXi doesn't have drivers for consumer GTX/RTX cards)

You definitely don't always need the enterprise GPUs for at least proxmox, I've seen plenty of people able to pass through stuff like 1060s.

9 minutes ago, Tad Bittoomuch said:

Also as far as I'm aware, it won't work at all with AMD gpus.

Proven to work with the newer 5xxx/xt gpus on proxmox as well as unraid.
I've seen a few people using the RX4xx/5xx GPUs on it as well.

[Tad Bittoomuch]

1 minute ago, pbx2 said:

You also definitely don't always need the enterprise GPUs for at least proxmox, I've seen plenty of people able to pass through stuff like 1060s.

Proven to work with the newer 5xxx/xt gpus on proxmox as well as unraid.
I've seen a few people using the RX4xx/5xx GPUs on it as well.

I will admit that I haven't played around with proxmox, but I was specifically referring to gpu passthoughs in ESXi

[pbx2]

3 minutes ago, Tad Bittoomuch said:

I will admit that I haven't played around with proxmox, but I was specifically referring to gpu passthoughs in ESXi

Seems that nvidia cards work on ESXi with some trial and error - at least they did ages ago
https://www.reddit.com/r/nvidia/comments/5m07h2/pci_passthrough_in_esxi_for_consumer_nvidia_cards/

[Shigeru]

7 minutes ago, son_of_ottie said:

Almost the whole working world is in that same boat, my friend. So not worrying is THE WAY to go. Use this time to do other stuff you wanna do...even if it's re-reading Harry Potter!

Yeah, in my case searching in my old tech collection to see what I can rebuild while I collect mental energy to do other things

[GOTSpectrum]

2 hours ago, GOTSpectrum said:

...

 

 

blog Update ETA --- Expected after 0500GMT ---

Currently working a second technical issue,

 

Blog ETA now no sooner than 0500GMT!

 

And yes, the whisky seal has been broken.

[gmouz]

29 minutes ago, TheAmazingDave said:

 

YAY!

 

Spoiler

Me too!

 

[The1Dickens]

I like on the stats page, it shows trajectories for overtaking. For example, LTT will overtake Team Hardware.no in 10.2 hours, and Team Folding@EVGA in 4.2 months. I assume that's a 'both teams operating at the current pace' calculation?

 

Anyways, my system seems to be chugging along just fine. Going to try to leave it up and running as long as possible.

[GOTSpectrum]

Just now, The1Dickens said:

I like on the stats page, it shows trajectories for overtaking. For example, LTT will overtake Team Hardware.no in 10.2 hours, and Team Folding@EVGA in 4.2 months. I assume that's a 'both teams operating at the current pace' calculation?

 

Anyways, my system seems to be chugging along just fine. Going to try to leave it up and running as long as possible.

Assume both teams keeping the current moving average, we are performing way above our moving average

 

[The1Dickens]

So, does a 'work server' both distribute and accept work units, versus the 2 'assignment servers' just pump out work units? Or do the 2 assignment servers pump work units to the work servers to be rationed out to the clients as possible? And what is 'CS' and linus1 is a 'NO'?

 

EDIT: I think 'CS' is collection server? So the linus1 collection server is down, it looks like?

[GOTSpectrum]

18 minutes ago, GOTSpectrum said:

Currently working a second technical issue,

 

Blog ETA now no sooner than 0500GMT!

 

And yes, the whisky seal has been broken.

 

Alright guys,

 

So... Technical issue one:  data sync issue. Resolved* 

 

Technical issue two: corrupted FAH data. Resolved.

 

*mostly resolved, some manual edit required. 

 

 

 

Datasheet update in progress, Blog ETA 0500GMT 

[GOTSpectrum]

1 minute ago, The1Dickens said:

So, does a 'work server' both distribute and accept work units, versus the 2 'assignment servers' just pump out work units? Or do the 2 assignment servers pump work units to the work servers to be rationed out to the clients as possible? And what is 'CS' and linus1 is a 'NO'?

 

EDIT: I think 'CS' is collection server? So the linus1 collection server is down, it looks like?

Assignment servers are the ones that tell the client it can have a WU from a WS

[Geralt]

Damn, the LTT server has 65.88TB dedicated for WUs. I wonder if there will be a video on the channel on how they set the server up.

[LAR_Systems]

2 minutes ago, Geralt said:

Damn, the LTT server has 65.88TB dedicated for WUs. I wonder if there will be a video on the channel on how they set the server up.

I love that covid and @GOTSpectrum created a folding team that is destroying F@H work servers, so Linus had to step in and help them deal with the folding team traffic that flys his flag!

[El_Psy_CoNGR00]

@Geralt where did you get the spec from? i am interested in looking at them for myself. 

 

[n0xlf]

I couldn't wait, so here's some AWS folding pr0n (depending on the exact project, the GPU's are doing about 32 million PPD, although it floats between 2M-4M per GPU so hard to say - In any case, it's 40,960 CUDA cores)

 

 

 

 

[Geralt]

7 minutes ago, El_Psy_CoNGR00 said:

@Geralt where did you get the spec from? i am interested in looking at them for myself. 

 

https://apps.foldingathome.org/serverstats

 

You can find the space used on this server stats page

[El_Psy_CoNGR00]

@Geralt  thanks

 

[The1Dickens]

11 minutes ago, n0xlf said:

I couldn't wait, so here's some AWS folding pr0n:

A.) How much does that cost, B.) Which CPU and GPUs are those, and C.) Do those fold better than the Titan V, which looks to have the same CUDA cores?

 

Also, would a Titan Z have been better (with 5760 CUDA cores versus 5120, it looks like?)?

Edited March 28, 2020 by The1Dickens
One more Q

[Gorgon]

7 minutes ago, n0xlf said:

I couldn't wait, so here's some AWS folding pr0n:

 

 

 

 

That's sick. I believe @Metallus97 did some work on getting the FAHClient Control Port (TCP 36330) working through the AWS firewall so you can use the Remote Access capability to monitor and control the AWS instances from a computer at home and consolidate the local and remote folding into one view. (See the Remote Access link in my sig)

[Walker1356]

I see the LTT intake server is up and running good job jakkuh!!