World Community Grid

I wonder how many people they're going to lose in the long term because of this.

It definitely sucks since it's really the only project I run outside of helping with some competitions and races.
 
I wonder how many people they're going to lose in the long term because of this.

It definitely sucks since it's really the only project I run outside of helping with some competitions and races.
Honestly, the users that care will still be around. The ones that don't probably won't even notice the boxes ran dry or went idle for a bit. The bigger issue will be how smooth of a transition it is after they open it back up to the public....
 
Honestly, the users that care will still be around. The ones that don't probably won't even notice the boxes ran dry or went idle for a bit. The bigger issue will be how smooth of a transition it is after they open it back up to the public....
The issue is that a lot of people will move on to something else to run because they don't want to be idle for two months. How many of those people are going to come back once they have something else setup and running? Obviously some will, maybe even a lot, but it won't be everyone simply because they might find something else they prefer running.

I simply don't understand how or why this would take two months. I could understand a week or so maybe but two months seems really odd.
 
The issue is that a lot of people will move on to something else to run because they don't want to be idle for two months. How many of those people are going to come back once they have something else setup and running? Obviously some will, maybe even a lot, but it won't be everyone simply because they might find something else they prefer running.

I simply don't understand how or why this would take two months. I could understand a week or so maybe but two months seems really odd.
Most likely involves moving/upgrading of hardware since it won't be hosted by IBM. They need to stress test it and make sure everything is good and I'm sure that can take a while to test.

As far as users, most of WCG isn't active users. Most are set and forget people. A lot of them probably won't even notice. The ones that are active most likely already support other projects as is. If not, I'm sure they will most likely not find another project that outweigh's WCG. Our project selection has really gone down the toilet over the last few years. Quite frankly, I don't see them leaving BOINC for another platform over a two month move either. So, odds are, their user base won't be affected much. I think what happens after the 2 months down time will be a bigger factor...
 
My plan is to load up this weekend with enough WCG work to last 7 days (Feb. 21st). When that work is finished, I'm going to switch one of my Windows machines over to FAH until WCG is back up again.

I stopped running the FAH Chrome client a long time ago when I started losing points. The standard FAH client appears to be much different now. Very similar look and feel to the retired Chrome client. I still have my passcode so I'm good there.

https://folding.extremeoverclocking.com/user_summary.php?s=&u=237271
 
Last edited:
I think those that remain with BOINC will merely attach another project rather than leave WCG altogether. This means they will remain but further contributions may be divided among the new projects as well. Something most active users do anyways.
 
My plan is to load up this weekend with enough WCG work to last 7 days (Feb. 21st). When that work is finished, I'm going to switch one of my Windows machines over to FAH until WCG is back up again.

I stopped running the FAH Chrome client a long time ago when I started losing points. The standard FAH client appears to be much different now. Very similar look and feel to the retired Chrome client. I still have my passcode so I'm good there.

https://folding.extremeoverclocking.com/user_summary.php?s=&u=237271
Any particular reason you chose FAH? Just curious.
 
1644691611295.png
 
As part of preparing for the final phase of WCG migration, we want to make volunteers aware of the following important dates:

  • World Community Grid pauses sending out work units: February 14, 2022.
  • World Community Grid pauses receiving work units: February 27, 2022.
  • Website and forum will be offline: from February 28 until April 22, 2022. A simplified version of the website will act as a sign post during the transition, to share updates and communicate our progress and roadmap for the World Community Grid in 2022.
During this final phase of the migration, we will post all updates and communications on the simplified WCG website, and will cross-post them on WCG Facebook and Twitter pages. Unfortunately, neither forums, nor the full WCG webpage material will be accessible during this time.
 
Any particular reason you chose FAH? Just curious.
Only interested in BioMed projects. I ran FAH for a couple years before switching over to WCG full-time. Already have a userid and passcode. Just need to reinstall the client software.
 
Last edited:
I don't care if it takes 1 month or three or whatever. What matters is its done right, and that there is no data loss. Serious contributors will have no issues finding other projects in the interim, and switching to or switching back is not hard. Casual contributors will likely not even notice the shutdown and will just keep clients going till work resumes. Massive server migrations take time to do properly and there is a lot at stake if you mess it up.
 
The issue is that a lot of people will move on to something else to run because they don't want to be idle for two months. How many of those people are going to come back once they have something else setup and running? Obviously some will, maybe even a lot, but it won't be everyone simply because they might find something else they prefer running.

I simply don't understand how or why this would take two months. I could understand a week or so maybe but two months seems really odd.
1644766129916.png
 
To build on to the screenshot above from Gilthanis, I'll give a little perspective from having worked in a data center for a number of years hand have been part of migrations like this. When you're moving servers, or moving data to bigger/updated servers, yes you can get that kocked out in days or weeks. But moving data to a new system, that's a completely different thing. Its not just putting your data files into the new database server, there's also translating the data into the new systems format. That alone can take days, or even more when it's a literal ton of data which I imagine WCG has. And these import/translate jobs aren't always the most graceful in handling things they weren't expecting. Maybe it manages to drop the row and proceeed, maybe it crashes. Maybe you have to have the entire table for the data to have meaning and can't have the unexpected record dropped. So as soon as you noticed the import stopped, you get to start reading through the logs, find the issue and fix it. Then probably start over from the beginning. Lather, rinse, repeat. Do we need all the historic data? Nobody is going to be the one to say no, so we better bring it all. Now its a month+ later and the data is finally imported into the new system. So we get to begin the process to validate it and then start testing the new system.

Yes, with time and resources you are able to do a lot of testing before hand and work out the kinks in the process. However, they seem to have not had sufficient time and resources to do enough testing so on to plan B. We have to figure this all out while we're doing it for real. And none of the above addresses all the politics that are going to come up along the way. People who aren't even stakeholders are going to have every opportunity to give very strong opinions on things they don't really understand and have nothing to do with them. Politics adds hours or even days to the project every time you turn around.

So while you may wonder how something like this could take 2 months, I wonder how they're going to get this done in only 2 months.

Edit: I haven't personally followed the discussion leading up to this, so the details may differ but the overall this is how this crap always goes is there.
 
Last edited:
And hello to everyone who may remember me, it's been a while. I don't have the resources available to contribute, but I do stop by and see what's up every once in a while. Right now all of my boxen are devoted to running VMs related to things I'm trying to fgure out for work, studying for certifications, and other general IT guy stuff. Maybe someday.
 
I'm glad you chimed in. It is always great to get a reminder that people still pay attention even if they aren't hammering away at work units. Sometimes the team seems smaller than it is and it is great to get reminders that people are still here in one form or another.
 
Schedule to stop generating new tasks for WCG projects can be found here. Grab OPN cpu and MCM tasks if you need to right now.
I am about to stop the processes that are loading work into BOINC. Once I do that, the projects will send out the remaining work that they have loaded into BOINC and then stop sending out new work. This will occur for the different projects as follows:

  • OpenPandemics - GPU: Is already out of new work
  • Help Stop TB and Africa Rainfall Project will stop within an hour
  • OpenPandemics - CPU and Mapping Cancer Markers will have work for about 9 hours

Note that we will send out resends in order to finish all of the work.
 
Periodically, WCG publishes updates on their main site regarding WCG and the migration. Last Fri's article has interesting stats on overall WCG computing power.

WCG in numbers - power


In this post we provide some behind the scene numbers about WCG

Published on: 11 Mar 2022

Have you ever wondered how big a computer should be to equate the World Community Grid computing power? For all of 2020, the average power of WCG was the equivalent computing power of a cluster running at 100% capacity 24x7 for 365 days with 11,999 nodes with 1 Intel Core i7-9700K CPU @ 3.60GHz. Indeed, GPUs could substantially increase this power, as shown in OPN1 March 8 post.
How is this amount distributed across volunteer device configurations? Table 1 shows the relative amount of computing power produced by computers with different RAM and CPU configurations. This calculation does not differentiate between CPU and GPU computing power, and about 35% of the computing power at the time of this report comes from GPUs.
a8_table_ram.jpg
Top 5 percentages are marked in BOLD (which together are 56.9% the power of WCG): 10.4% of our computing power comes from computers with 64GB or more RAM and 32 or more cores, 11.9% from computers with 32GB-63GB RAM and 16-31 cores, 16.4% from computers with 16GB-31GB RAM and 8-15 cores and 10.1% from computers with 8GB-15GB RAM and 4-7 cores.

What platforms are more frequently used? 71.11% of the computing power is provided by Windows OS with X86/64 processors, 25.51% by Linux OS (25.06% with X86/64 processor and 0.45% with ARM processor), 0.01% by FreeBSD OS with X86/64 processor, 2.83% by Mac OS (2.78% with X86/64 processor and 0.05% with M1 processor) and 0.54% by Android OS with ARM processor.
 
Krembil is currently doing WCG testing and so far appears to run smoothly . They will provide an update next week after doing the test on both the staging and production environments. Hopefully WCG will be back sometime in April. Full post can be found here or below.

Streaks Will Be Extended to Cover Downtime, Website and Forums Have Been Restored in Krembil Staging Environment


All streaks that were active as of February 16th, 2022 will be extended to cover the full downtime plus a grace period. Website and Forums are deployed in Krembil staging environment, testing is underway.

Published on: 25 March 2022

We are happy to report that the website and forums have been deployed in the Krembil staging environment. Access to the website and forums for volunteers will be possible pending a successful testing phase in both staging and production environments. We will provide an update next week on the outcome.
Streaks have been preserved and will be extended to span the elapsed downtime once we resume sending workunits to volunteers. All active streaks as of February 16th, 2022 will automatically cover all downtime, plus a grace period of 10 days to allow for the flow of workunits to resume in full.
We have now validated several components of the World Community Grid workflow via integration testing. To do so we first provisioned a remote server to mimic the process of obtaining input batches from our research partners, uploading batches for each project to the server. Our backend successfully connected to the remote server, indexed the available batches, downloaded all new batches, and verified their integrity. The successful completion of these tests validates correct configuration and coordination of our workers, databases, and scheduler.
Following the completion of these steps and the previous stress test of our storage layer, we are now preparing for the final system test.
We thank you for your support!
Krembil WCG Team
 
Looks like WCG will not be included in the upcoming pentathlon unless the organizer skips real testing by contributors for potential bugs and take the risk. It's a nice project to run though.
 
Well....I'm kind of glad at the same time as sad because I'm sure someone in the Pentathlon committee is probably eyeballing some way of having them in the event...lol. I for one would not want that to happen with them just bringing things back up as all.
 
WCG Restart Update

Restart update


WCG is almost ready to restart.

Published on: 22 May 2022

QA testing has finally successfully finished and all the bugs have been resolved. Production environment is being tested right now. Considering the unexpected issues we ran into with the QA system, we prefer to test it a couple of days further. The website and forum are ready to be relaunched, the production environment is successfully querying BOINC locally and we are going to restart BOINC slowly in the next few days. We will provide further details early next week.
a_0522.png

Thank you
WCG Tech team
 
https://www.worldcommunitygrid.org/news/0603

They did not provide a new restart date. I guess they learned their lesson. They're struggling to complete the migration and they're working on improving communications? Their #1 priority should be getting WCG operational again. This doesn't look good.
I would prefer that they scrap the Frankensteined setup and just update to the latest BOINC server. Then work on getting all the other controls they want added later. Maybe actually help develop the server code to improve all projects. But, you know that will never happen.
 
Today's update on WCG migration status: https://www.worldcommunitygrid.org/about_us/article.s?articleId=773

Some bumps along the road. Hoping that they can resolve gpu workunit issue. I'm patiently waiting for it. ;)

July 2022 Tech Team Update​


Discusses current errors with our BOINC platform and what is currently being fixed


Published on: 13 Jul 2022



Dear volunteers,
Over the past week, we entered a new stage of our testing phase where we gave a few workunits to our volunteers to crunch. The goal was to see if any unforeseen errors occurred, which we were able to find thanks to the feedback of our volunteers. For specific questions you may have on the limited workunits, check this thread on our forums.
Here’s a quick update from our tech team:
We have been working to assess how well the system responds to load and whether configuration is correct. As many have noted on the forums, there are some issues with the BOINC backend configuration. Users will have noticed off-putting notices in the BOINC manager about incorrect project URLs and attaching twice to the same project. We have addressed the underlying issue and apologize for the inconvenience.

In addition, we are working to resolve the lack of synchronization between the website profile page and the BOINC manager. We are also working to fix a blocking issue with the GPU workunits for Open Pandemics. Once this is resolved, we should be able to increase the quantity of workunits we can send out to volunteers.

Our backlog of non-blocking issues has continued to grow as we observe load on the system, but overall we are nearing a stable state of affairs and look forward to updating users with statistics and more details as we continue the roll out of our production system.
Thank you for your understanding and continued support.
Sincerely,
The World Community Grid team
 
I have an Odroid N2+ and a N2 running, and neither have got any of the few WCG tasks sent out so far. Maybe it has been all x86 so far?

Combined they chew about 14W running full out and I am willing to sweat for that :)
 
Back
Top