Question Is DC dying out?

Traxan

Senior member
Jun 5, 2005
374
8
81
I'd like to think I'm wrong but it looks like distributed computing is falling way out of favor.

- The major DC page hasn't been updated in 3 years
- Folding@Home has all but vanished
- The World Community Grid page is broken ad out of date
- Very little discussion here.

Thoughts?
 

VirtualLarry

No Lifer
Aug 25, 2001
56,443
10,113
126
"Traditional" DC has largely failed on it's promise, in many respects. And thus, it is getting replaced by Web3 distributed compute / storage / web / finance projects.

Plus, time is money. Paying for super-computer time is more feasible these days.
 
Last edited:

Markfw

Moderator Emeritus, Elite Member
May 16, 2002
25,738
14,770
136
I'd like to think I'm wrong but it looks like distributed computing is falling way out of favor.

- The major DC page hasn't been updated in 3 years
- Folding@Home has all but vanished
- The World Community Grid page is broken ad out of date
- Very little discussion here.

Thoughts?
Major DC page ? The posts on this forum are updated almost daily.

Folding@home vanished ? I don't get that, I do 80 million ppd, and we just finished a competition.

Yes, WCG went down 4 days ago, no ETA, but they are working it, and it will be up before too long.

Most other projects are running along just fine, and we just finished a primegrid competition.

Very little discussion here ? updated daily.
 
Reactions: Fardringle

Kiska

Golden Member
Apr 4, 2012
1,024
291
136
Plus, time is money. Paying for super-computer time is more feasible these days.

I don't think that is quite true. Most institutions supercomputer time is governed by how they allocate time to projects. For example at my university we have a modest supercomputer its 1.5PFLOP/s peak. Currently for a PhD student they get told they can have 24 hours of shared compute every 7-9 weeks, every institution/organisation/country/etc is different.
 

Kiska

Golden Member
Apr 4, 2012
1,024
291
136
82.58 TFLOPS for 4090 card.

Requires 18 4090 cards to match that.

But at the same time that is if you're doing FP32, boy if you do FP64 say goodbye to that max performance

But yes to the original thoughts, every year more people leave DC, but traditionally we had better CPUs to cover the "shortfall"
 

Markfw

Moderator Emeritus, Elite Member
May 16, 2002
25,738
14,770
136
But at the same time that is if you're doing FP32, boy if you do FP64 say goodbye to that max performance

But yes to the original thoughts, every year more people leave DC, but traditionally we had better CPUs to cover the "shortfall"
My own farm has gone ballistic in power compared to 10 years ago. The power bill has gone from $200 to over $1000 and they are more efficient than ever. Gone from a handful of cores to over 1800 threads.
 

Skillz

Senior member
Feb 14, 2014
946
979
136
I'd like to think I'm wrong but it looks like distributed computing is falling way out of favor.

- The major DC page hasn't been updated in 3 years

Guessing you mean this page: https://teamanandtech.com/
I don't know who runs that site, but the other site is www.teamanandtech.org which has been getting less and less use recently due to Discord. It's still got relevant information posted, but most people just need it to re-read up on the information and don't necessarily have to post anymore.

- Folding@Home has all but vanished

Can you elaborate on this?

- The World Community Grid page is broken ad out of date

What WCG Page?

- Very little discussion here.

We tend to only post here on Sundays when the weekly stats are posted or during a competition.

Speaking of competitions this is a competition site hosted/ran by members of TAAT.

But most of our discussions happen on Discord: https://discord.gg/E5z65jR5Ka

Thoughts?

Yes, need more input from you.
 
Reactions: Fardringle

StefanR5R

Elite Member
Dec 10, 2016
5,682
8,240
136
Rumors of DC's death are probably exaggerated. Perhaps there is some sort of stagnation though, or even some decline.

There are two sides to that:

1.) From the scientific side, only problems of which the solution can be partitioned into a tremendous number of tiny parts without (or only very little) interdependence are suitable for DC — this always was and will remain true.

While you could move such problems from DC to supercomputers right away, that would be a waste of the extraordinary interconnect power of supercomputers. And since there practically are no supercomputers with idle time, such moves won't happen.

The more realistic alternative to classic volunteers based DC would be DC in the cloud. It certainly is true that cloud computing has become much more accessible and affordable over time. But still, the costs are not at all negligible. Most scientific projects have to fight a lot for getting a budget, and if they luck out and get one, they have to spend it wisely.

It is true that available computing capacity in volunteer based DC cannot be controlled by the project owners. However,
– hobbyist projects basically have all the time in the world,
– a few platforms exist which have been hosting many different sub-projects for a long time now (longer lasting projects or/ and shorter lasting ones) and therefore have a good grasp on their available computing capacity, and thus on the planning of the computational part of the projects. Not just WCG but actually F@H, Rosetta@home, and CPDN are such platforms.

2.) From the volunteers' side, one related question would be: Is desktop computing dying out? Of course, several DC projects adopted mobile computing too, including computing on telephones. But let's face it, the form factors imply that desktop computers and occasional compute servers are doing the bulk of the work in DC.

As others already mentioned, the cost of electricity has obviously become more of a concern for the typical DC volunteer demographic in recent years, not to mention most recent years and months. I suspect this does not only cut into how many households take part in volunteer based DC, but it especially might cause fewer and fewer corporations and publicly funded organizations to be willing to donate computer time to DC.
 
Last edited:

Fardringle

Diamond Member
Oct 23, 2000
9,190
755
126
Most institutions supercomputer time is governed by how they allocate time to projects. For example at my university we have a modest supercomputer its 1.5PFLOP/s peak. Currently for a PhD student they get told they can have 24 hours of shared compute every 7-9 weeks, every institution/organisation/country/etc is different.

My own farm has gone ballistic in power compared to 10 years ago. The power bill has gone from $200 to over $1000 and they are more efficient than ever. Gone from a handful of cores to over 1800 threads.

It would be fascinating (to me) to see how some of the individual "supercomputers" on this team compare to the one at Kiska's school!
 

Markfw

Moderator Emeritus, Elite Member
May 16, 2002
25,738
14,770
136
It would be fascinating (to me) to see how some of the individual "supercomputers" on this team compare to the one at Kiska's school!
The two biggest computers (in terms of threads) that I have is dual 7763 Milans and dual 7V12 Romes. They are 256 threads each, so there is 512 of my threads right there. I would not call them supercomputers. Want to see a picture of them ? and 3 7742 Milans is 384 more, so 5 computers is 900 threads. Then 8 5950x and 4 7950x is 394 more. one 64 thread Rome, and 5 64 thread Naples is 384 more. Still no supercomputers., a 2683v3, a 2970wx a 1950x.. I think 1800 is a little high... more like 1400.
 
Last edited:

Fardringle

Diamond Member
Oct 23, 2000
9,190
755
126
The two biggest computers (in terms of threads) that I have is dual 7763 Milans and dual 7V12 Romes. They are 256 threads each, so there is 512 of my threads right there. I would not call them supercomputers. Want to see a picture of them ?
I was thinking more of a comparison in total computing power. Kiska said the school system peaks at about 1.5PFLOPS.

According to Google, your dual 7763 is capable of about 7.16TFLOPS, but I'm not finding any numbers for the 7V12 Romes...

Add in all of the GPUs and the numbers go up really fast!

My puny collection of CPUs is about 8.3TFLOPS total (per Google), and my single RTX 3060Ti by itself can produce about 16.2TFLOPS.
 

Markfw

Moderator Emeritus, Elite Member
May 16, 2002
25,738
14,770
136
I was thinking more of a comparison in total computing power. Kiska said the school system peaks at about 1.5PFLOPS.

According to Google, your dual 7763 is capable of about 7.16TFLOPS, but I'm not finding any numbers for the 7V12 Romes...

Add in all of the GPUs and the numbers go up really fast!

My puny collection of CPUs is about 8.3TFLOPS total (per Google), and my single RTX 3060Ti by itself can produce about 16.2TFLOPS.
I saw a table here: https://www.geeks3d.com/20140305/amd-radeon-and-nvidia-geforce-fp32-fp64-gflops-table-computing/

And it says that 16.2 is gflops, but its fp32, and it also has fp64 rating. The same can probably be said for CPUs. My 4090 and 3090TIs are not even listed, but at least my 4 Titan V's are. It would make my head spin to try and figure this out.

But I found that my 5950x's are 190 gflops each or 1.5 Pflops alone, let along the 7950x's at 242 or 968 for 4 of them. Thats almost a pflop right there. Then the 7742's, the dual 7v12 (just a little higher than 7742's) and the 2 7763s

Edit: after looking closely, I am not sure all these sites are defining tflops and gflops the same way.7.16 tfops for 256 cores ? and 5950x's are 190 gflops each ? (https://boinc.bakerlab.org/rosetta/cpu_list.php)

also, my math may be off. 1000 gflops = 1 tflop ? and 1000 tflops = 1 pflop ???
 
Last edited:
Reactions: Fardringle

Markfw

Moderator Emeritus, Elite Member
May 16, 2002
25,738
14,770
136
I just mentioned in the CPU forum, something that may be affecting DC work. Someone who does as much as me would see a $2,707 monthly electric bill !! That will cut down your DC work....
 
Reactions: Fardringle

shortylickens

No Lifer
Jul 15, 2003
82,854
17,365
136
People with spare processing power are now using to mine crypto. Even though the electricity requirement means you are usually losing money. My electric bill keeps increasing. If I'm not using my system I put it in standby. And as pointed out earlier, if a company or university or whatever needs a large amount of work done, they just pay for a supercomputer or server farm. They cannot rely on strangers around the world.
 

cellarnoise

Senior member
Mar 22, 2017
728
399
136
People with spare processing power are now using to mine crypto. Even though the electricity requirement means you are usually losing money. My electric bill keeps increasing. If I'm not using my system I put it in standby. And as pointed out earlier, if a company or university or whatever needs a large amount of work done, they just pay for a supercomputer or server farm. They cannot rely on strangers around the world.
There are still many projects that are thankful for D.C. contributors. And I have been in contact with others that are thinking about it. Regional electric power cost on most of the planet to varying degrees is a deterrent trend for many, but, under certain circumstances is also driver for researchers to look to D.C. once again.

Planetary fluctuations are not unnoticed from researchers.
 

Icecold

Golden Member
Nov 15, 2004
1,090
1,008
146
F@H should be paying you!
apparently, that's exactly what curecoin (I think that's it) used to do !
Folding@home is now whitelisted on Gridcoin, and you can still run Folding@home as part of Team Anandtech while earning Gridcoin, but it requires creating a new user account since you have to put your Gridcoin address or something like that in your user name for them to be able to track it.
 

StefanR5R

Elite Member
Dec 10, 2016
5,682
8,240
136
Just pointing out a little detail: Cryptotoken miners are not paid. They are merely given cryptotokens for their work. They can then take part in the cryptotoken snowball system and try to make money off it.
 
Reactions: Ken g6

shortylickens

No Lifer
Jul 15, 2003
82,854
17,365
136
Just pointing out a little detail: Cryptotoken miners are not paid. They are merely given cryptotokens for their work. They can then take part in the cryptotoken snowball system and try to make money off it.
Yeah thats what I meant, not a paycheck, but a chance at a currency that probably wont make them any money and most likely cost them money. But because so many people today cant do basic math, its still very popular.
 

Assimilator1

Elite Member
Nov 4, 1999
24,120
507
126
I'd like to think I'm wrong but it looks like distributed computing is falling way out of favor.

- The major DC page hasn't been updated in 3 years
- Folding@Home has all but vanished
- The World Community Grid page is broken ad out of date
- Very little discussion here.

Thoughts?
1. Which one?
2. Nope! Why on earth did you think that?
3. Yes, a temporary situation which recently occurred and repairs are underway. If you'd had a quick look through the DC forum you would've seen my thread called 'WCG problems' where it's being discussed .
4. Less discussion here than 10yrs ago or longer, yea that's true. I think in part that's down to DC being more reliable (mostly, lol), but I think perhaps many people spend less time in forums than they did? (I know I do, mostly!).
I'm a member at the M3post forum, and it's most active period was some years ago (although it's still plenty active). I'm also an admin at a gerbil forum (the main English speaking one AFAIK), peak activity there was around 2008-2013 and then it dropped off, I think it levelled off and stayed steady a few years after then though.

I wonder how activity across the rest of AT's forums compares? And other forums? I know OcUK's DC forum is very quiet now, but I don't know about their other sections.
 

Ken g6

Programming Moderator, Elite Member
Moderator
Dec 11, 1999
16,282
3,903
75
apparently, that's exactly what curecoin (I think that's it) used to do !
Also FoldingCoin.
Just pointing out a little detail: Cryptotoken miners are not paid. They are merely given cryptotokens for their work. They can then take part in the cryptotoken snowball system and try to make money off it.
Right. I've been trying to get paid for DC since the start. FLDC made me about tree fiddy in BTC dust. Still haven't converted it either, so I've really made nothing!
 

VirtualLarry

No Lifer
Aug 25, 2001
56,443
10,113
126
Right. I've been trying to get paid for DC since the start. FLDC made me about tree fiddy in BTC dust. Still haven't converted it either, so I've really made nothing!
Next-gen PoW coins, using PoUW, will solve DC problems, and you can get paid.

Dynex coin is a leader right now in that.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |