Brainstorming BOINC Projects #008
What is BOINC?
BOINC is an open-source volunteer oriented distributed computing grid. It's free, and harnesses the unused clock cycles from processors and graphics cards to attempt to cure cancer/AIDS/Ebola/malaria, map the Milky Way, fold proteins, search for extraterrestrial life, etc.
Currently, there are approximately 176,000 active BOINC users & over 4.5 million registered accounts. There's serious computing power within the BOINC community and the vast majority of that work goes unpaid.
What are some examples of existing BOINC projects?
- World Community Grid : Attempting to solve cancer, Ebola, malaria, Zika, etc.
- SETI@Home : Search for Extraterrestrial Intelligence (SETI).
- Milkyway@home : Creation of a 3D map of the Milky Way galaxy using data gathered by the Sloan Digital Sky Survey. This project enables research in both astroinformatics and computer science.
Examples of more Gridcoin-whitelisted projects.
How does Gridcoin benefit BOINC?
The Gridcoin network rewards BOINC volunteers for the work done on whitelisted BOINC projects. As a result, more computation is pointed towards their projects without costing them a penny...
By comparison, a project administrator would have to fork out serious cash (pay-to-play) to get an equal level of computing power out of Maidsafe/Golem/etc. These platforms also limited in their possible types of computation. Through BOINC you can distribute any kind of distributed computing application.
Who can make a BOINC project?
Anyone can create a BOINC project, as there is no centralized authority in charge of the creation of BOINC projects.
How can I make a BOINC project?
One of the easiest method of creating a BOINC project is to use the 'BOINC Server Docker' repo!
The developer "Marius Millea" (Cosmology@Home dev) is very active, give him a shout if you run into issues w/ the docker container.
Outdated guides
- 'Creating BOINC projects (PDF)'
- 'Project creation cookbook'
- 'Building BOINC and BOINC applications on Linux'
There are three types of BOINC projects!
Volunteer/Non-Profit
- Scientists: Use BOINC to create a volunteer computing project.
- Universities: Use BOINC to create a Virtual Campus Supercomputing Center.
The majority of BOINC projects are volunteer/non-profit based projects.
Commercial/For-Profit
- Companies: Use BOINC for desktop Grid computing.
Create a commercial BOINC project, charge customer to process their data and distribute x% to BOINC crunchers & y% to project admins/shareholders.
Existing research (mostly storage decentralization, not 100% decentralized)
- Optimizing the data distribution layer of BOINC with BitTorrent.
- Attic: A Case Study for Distributing Data in BOINC Projects.
- Extending the BOINC architecture using peer-to-peer application code exchange.
BOINC vs Commercial Cloud Computing
Research papers
- The cost of virtualization implementation in volunteer computing
- Increasing GP Computing Power for Free via Desktop GRID Computing and Virtualization
- [PDF] Cost-Benefit Analysis of Cloud Computing versus Desktop Grids
Research TL;DR: (Note - 8 year old statistics!)
- In the best-case scenario, hosts register at a rate of 124 cloud nodes per day.
- The ratio of volunteer nodes needed to achieve the compute power of a small EC2 instance is about 2.83 active volunteer hosts to 1.
- Effectively, with a commercial BOINC project, there exists the opportunity to create a cloud computing service without actually owning/providing the computing capabilities yourself. Imagine creating Azure at a fraction of the cost..
Got an idea for a BOINC project?
Do you have an idea for a potential BOINC project, or do you have questions about either BOINC or Gridcoin? Reply in the comments below!
You don't have to know how to code or anything like that. This thread is for brainstorming ideas. Put your ideas out there, and maybe someone else will come along and make it for you!
By posting an idea to this thread, you agree to open sourcing the idea under the MIT license within the BOINC brainstorming thread repo.
Why is this thread being reposted?
It's been a while. Perhaps we've all come up with some new and different ideas that are worth discussing. :)
Past threads
#007
#006
#005
#004
#003
#002
#001
That is all that I can think of at the moment
The Big and Ugly Rendering Project (B.U.R.P.) http://burp.renderfarming.net/ is a BOINC project that offers a public render farm for Blender animations.
A couple of years ago BURP was used to re-render the Blender Foundation's short film Big buck Bunny (codenamed Peach Project) in 4K resolution at 60fps and with stereo vision.
Visit: http://bbb3d.renderfarming.net/ to view or download the results.
Things are moving a bit slowly at BURP right now which is partly due to a lack of input from content creators.
Anyone with an interest in distributed rendering should check it out.
Distributed rendering should be a great feature !!! I will check this BOINC project. I think that if it's working and usable by everybody a poll to whitelist this type of project could make the thing ! Thanks for the info.
Unfortunately BURP needs more session contributors and perhaps some more development before it is ready to be re-listed with Gridcoin. A couple of big sessions can keep the render farm busy for months but there have only been a handful of big sessions recently resulting in extended work unit shortages.
It is best suited to very detailed renders that would take months or years to render on one machine, the session vetting process takes a long time so it's not necessarily a good solution if you just want to off-load your own rendering resources unless you have a very flexible timeline.
It's a great project and definitely worth a look if you are into Blender rendering.
Next week I had an idea with a friend making music. Why not a project to share computing power with theese sound scientists using a lot of .vst (plugins) in their DAW's projects and not having enough ressources for that...?
I think it's a commercial one and as I don't have enough knowledge in GRC science I don't know if it would be possible to do but now the idea is here =)
that's a really cool idea. i was wondering if the same thing could be pulled off for gaming... but latency would be an issue.... never thought about using the power for DAWs. this is genius. latency wouldn't be an issue if it were just to provide minimum system requirements, etc. :D
one problem we might run into is copyright stuff. maybe we could convince some of the better VST producers to host a few for free as proof of concept?
Yes, we could convince some of them. Another solution is to make our free-distributed-computing DAW but it's another work ^^ We could use open source VSTs to test it too, and then try to convince the more mainstream ones with that.
more on this matter in this document: people.eng.unimelb.edu.au/vkostakos/files/papers/percom17.pdf
re: offloading tasks in an IoT environment. i've dropped a line to the people at ubuntu-studio, we'll see what transpires. :)
One of my ideas was the Open Sensor Array.
It is to be an Arduino-based opt-in sensor system, sensing RF, WiFi, light, dust, possibly all weather, noise, etc... and putting those results against health authorities results for depression, road rage, murder, suicide... to try and see if there is any correlation between the waves and particles flying through our bodies, and our moods.
It sounds a bit like hippie science, I know, but I feel at this point it's barely even being looked into.
The reason I haven't started this project yet was due to sensor-spoofing concerns. I have been doing some reading regarding this, and am open to any suggestions as to how to verify open-source results from across the planet.
Interesting reading on the matter: https://www.usenix.org/conference/woot16/workshop-program/presentation/park
Also a concern, is the whole Big Brother scenario. This sensor system will be admittedly invasive.
Also a concern, is the whole Big Brother scenario. This sensor system will be admittedly invasive.
The funny thing is that such a system is already in operation. Mobile operators / smartphone producers / app developers constantly spy on the users, with or without our permission.
i don't see how this is funny. i'm sorry.
and you kind of unwittingly made my main point: that we could create a dataset that rivals Google's etc, but it would be a freely available pool of information. information is valuable in today's world, and this could help change the value of said data. hide in plain sight, as it were.
Don't get me wrong, I'm not laughing. Maybe "funny" was the wrong word - more like "interesting" would be more suitable.
Anyway, I doubt megacorps like Google would allow creation of such a project. Sorry, but the truth is it would hit them pretty hard, so it's obvious they would do everything in their power to stop this.
all the more reason to try. :)
Another idea I had recently while talking with a gentleman at a Christmas party, was to develop a compression-testing network for OTT (over-the-top) content delivery.
The companies themselves run testing through GPUs... they have to put the content on a central distribution node, then compress that into smaller packets, and deliver those packets. Maybe BOINC can help "perfect" this compression+delivery technology.
Anyone paying attention to the Net Neutrality fiasco, by now knows that the whole thing is just so the cable companies can broadcast their unwanted material through a portion of our (the end user) apportioned bandwidth.
Since money runs everything... this seems an inevitability, and nerds like us can't help but ease the brunt of it all. :D
I personally have done WorldWide Grid Project BOINC, for more than 5 years, and my computers run 24/7 for this purpose mostly. I see that if new projects that help all peoples of this earth then yes, let it be so, but if it is just a new venture to make something that will eventually profit a few then go to OpenSource, that is the venue for that!
nice photo
Thanks for the lovely content
Interesting... i would like to gather more info about that