PDA

View Full Version : What factors do you consider when allocating your resources?



litehouse43
12-13-12, 12:18 AM
I recently started as a volunteer admin for a BOINC project (NumberFields@Home) and the endeavor of gaining new members from both within the existing BOINC community and from outside has gotten me thinking about my own allocation of resources. I am curious to hear from others how you choose who gets access to your valuable computational resources. What do you value most? I think personally I look primarily for a project that communicates well and is doing interesting science. How important are things like website layout, screensavers, points, badges and the like to everyone else? Do you care if the project is domestic or international? I would love to get input from other crunchers both for my own curiosity as well as to aid my efforts in continually improving the projects I am associated with.

pinhodecarlos
12-13-12, 03:23 AM
Can I ask you litehouse43 if you have a math degree? Before being a volunteer admin of any kind of project I think we must study (read articles, papers, etc) the matter to see if real science/math is made and not be a volunteer admin to gain new members, to be visible, etc...

When I look to run a project I only do the math ones and the reasonable ones. I read papers, master thesis on the subject, I try to understand the algorithm behind the application and try ways to tweak it. For me this is what makes a real cruncher, the one who knows the state-of-art in all aspects.

I don't care if project is domestic or international, I don't care about badges, I don't care about website layout, I care about the subject we are testing it, the rest is secondary. I don't care if the project only has 100 members, these ones at the end of the run have the grant of having done something special, something that is publishable. I'm on math projects to have my name on papers as I have on a few from OddPerfect project and others math related.

Sorry for my post but that is what I think.

Carlos

Fogle
12-13-12, 07:03 AM
Hi thanks for your question. For me I started like many others on SETI project years ago, it was intresting but I lost intrest rather quickly as looking for a needle in a haystack isn't much fun. I like to take a top level look at the science as I'm not a scientist. I like the space projects like Milkyway. Lately though I've been chasing milestone in every project I can find. Tuff job as there are many of them. Also it seems team racing and challanges is taking more and more of my time. Not complaining as this seems to be the most fun. So to sum it up I believe all the points you mentioned are valid as people's intrest change as time goes by. Keep crunching!! :cool: Go Team SETI.USA!! :-bd

kmanley57
12-13-12, 10:29 PM
I am much like Fogle, except I am not really chasing any Milestones. Just here to help people, and to find E.T.! :D

Quit looking for any depth of understanding math, besides not being in the RED financially. Science that was back when I was young, not so much anymore.

Mike029
12-14-12, 08:53 AM
I really liked Orbit@home and intelligence realm. PG, Seti, Einstein as well. Now more of a credit/challenge nut I guess. The more credits the more I like a project. :o

Duke of Buckingham
12-14-12, 09:19 AM
:p Everyone looking for the ET and I am here, I found you first. :D

:) http://www.toyart.com.pt/15603-large/dress-up-et-the-extra-terrestrial-series-1-action-figure-13-cm.jpg :)

Mike029
12-14-12, 09:31 AM
:p Everyone looking for the ET and I am here, I found you first. :D

:) http://www.toyart.com.pt/15603-large/dress-up-et-the-extra-terrestrial-series-1-action-figure-13-cm.jpg :)

=))=))

nanoprobe
12-14-12, 11:27 AM
I started out at WCG in 2008. Since I lost my mom to cancer and I am a cancer survivor finding cures for it and other diseases is where my heart is. Now that WCG has a GPU app I have all my resources there for now. I also run other medical related projects like POEM, SIMAP, Docking and GPUGrid when work on WCG is not available. I also like exploring our universe type projects so I do occasionally wander over to Einstein and Milkyway.
Badges, points, credits, whatever are nice but results are what I'm looking for. Over 1,000,000 results returned @ WCG so far. Hopefully that puts the project scientists that much closer to cures.

Al
12-14-12, 11:45 AM
I started out at WCG in 2008. Since I lost my mom to cancer and I am a cancer survivor finding cures for it and other diseases is where my heart is. Now that WCG has a GPU app I have all my resources there for now. I also run other medical related projects like POEM, SIMAP, Docking and GPUGrid when work on WCG is not available. I also like exploring our universe type projects so I do occasionally wander over to Einstein and Milkyway.
Badges, points, credits, whatever are nice but results are what I'm looking for. Over 1,000,000 results returned @ WCG so far. Hopefully that puts the project scientists that much closer to cures.

My sentiments exactly.

pinhodecarlos
12-14-12, 11:54 AM
I had a friend aged my father who died in 1999 with lung cancer so in that time I joined United Devices Project to crunch wu's that supposedly would find a cure for cancer. After 2 years or more running it, also as age was passing through me, and also as I started to study at university I interior found and understood that the cure for cancer is not achieved by simulation but on a lab making experiments. Since then I moved to math related projects.

Carlos

Fire$torm
12-14-12, 01:33 PM
Hello litehouse43,

Apologies for not posting sooner...

Although I am competitive by nature and gravitate to project challenges and raids, I've always been very interested in technology. Most any kind of tech, which is one of the reasons I took up computers as a hobby. So I like projects (in general terms) that advance man's pool of knowledge. But I prefer projects that give good credit and scale credit according to the method of processing, CPU vs. GPU, and the amount of time it take to complete a task. The concept of "Normalizing" credit across all BOINC projects is in my opinion not a good thing. Others may fell differently.

Anyhow, thank you for taking the time to ask. It is a rare thing to see a project admin taking interest in the personal side of volunteer computing, at least in this manner.

Happy Holidays,
Fire$torm

litehouse43
12-16-12, 04:47 PM
Can I ask you litehouse43 if you have a math degree? Before being a volunteer admin of any kind of project I think we must study (read articles, papers, etc) the matter to see if real science/math is made and not be a volunteer admin to gain new members, to be visible, etc...

When I look to run a project I only do the math ones and the reasonable ones. I read papers, master thesis on the subject, I try to understand the algorithm behind the application and try ways to tweak it. For me this is what makes a real cruncher, the one who knows the state-of-art in all aspects.

I don't care if project is domestic or international, I don't care about badges, I don't care about website layout, I care about the subject we are testing it, the rest is secondary. I don't care if the project only has 100 members, these ones at the end of the run have the grant of having done something special, something that is publishable. I'm on math projects to have my name on papers as I have on a few from OddPerfect project and others math related.

Sorry for my post but that is what I think.

Carlos

I actually do not have a degree in math; my formal education is in computer science and aerospace engineering. With that being said I am passionate about math however.


Hello litehouse43,

Apologies for not posting sooner...

Although I am competitive by nature and gravitate to project challenges and raids, I've always been very interested in technology. Most any kind of tech, which is one of the reasons I took up computers as a hobby. So I like projects (in general terms) that advance man's pool of knowledge. But I prefer projects that give good credit and scale credit according to the method of processing, CPU vs. GPU, and the amount of time it take to complete a task. The concept of "Normalizing" credit across all BOINC projects is in my opinion not a good thing. Others may fell differently.

Anyhow, thank you for taking the time to ask. It is a rare thing to see a project admin taking interest in the personal side of volunteer computing, at least in this manner.

Happy Holidays,
Fire$torm

I think that hits the nail on the head for my motivations. I wish more projects would take the time to see why people contribute and make the experience more mutually beneficial. I got involved in volunteer computing, like everyone else, to contribute to science and building the knowledge base of humanity.

I think sometimes projects are so concerned about the scientific output that they forget about the human side of volunteer computing. People want to get something out of these projects; they want to know their resources are being used efficiently. Volunteer computing is also a way to introduce other aspects of society, like marketing, into scientific exploration to help make science more mainstream.
From an administrative perspective there are two main issues. The first is getting volunteers already involved in BOINC to participate in your project while the second is getting people outside the BOINC community to come on in and check it out. These challenges are accomplished in different ways but I feel like it is something we all should always be thinking about. I know I am.

I was hoping some of the more competitive minded SETI.USA members could point me in the right direction for how to start project challenges. There is very little documentation provided on this topic from either BOINC or any supporting applications. I think starting some challenges may be a way to help grow our project internally, test our servers and hopefully gain more volunteers. Overall it seems to be something volunteers like to do. I personally have never really done any so I am at a loss for how to proceed on this end.

Thanks for the input everyone, I was hoping to get a little discussion going on the topic and am glad everyone seemed to be interested.

denim
12-17-12, 01:43 PM
For me, it depends. I could be helping on a challenge, trying to reach a milestone or wanting to get a warm and fuzzy where credits don't matter just the thought of what it is going towards.

Some are defunct or very intermittent with making WU's available, but a few projects I stay on regardless.

Fire$torm
12-17-12, 08:18 PM
I actually do not have a degree in math; my formal education is in computer science and aerospace engineering. With that being said I am passionate about math however.




I think that hits the nail on the head for my motivations. I wish more projects would take the time to see why people contribute and make the experience more mutually beneficial. I got involved in volunteer computing, like everyone else, to contribute to science and building the knowledge base of humanity.

I think sometimes projects are so concerned about the scientific output that they forget about the human side of volunteer computing. People want to get something out of these projects; they want to know their resources are being used efficiently. Volunteer computing is also a way to introduce other aspects of society, like marketing, into scientific exploration to help make science more mainstream.
From an administrative perspective there are two main issues. The first is getting volunteers already involved in BOINC to participate in your project while the second is getting people outside the BOINC community to come on in and check it out. These challenges are accomplished in different ways but I feel like it is something we all should always be thinking about. I know I am.

I was hoping some of the more competitive minded SETI.USA members could point me in the right direction for how to start project challenges. There is very little documentation provided on this topic from either BOINC or any supporting applications. I think starting some challenges may be a way to help grow our project internally, test our servers and hopefully gain more volunteers. Overall it seems to be something volunteers like to do. I personally have never really done any so I am at a loss for how to proceed on this end.

Thanks for the input everyone, I was hoping to get a little discussion going on the topic and am glad everyone seemed to be interested.

The easiest way to get a challenge noticed is to issue a "Project Challenge" on the BOINCstats site.

A few things to keep in mind...

*Project server(s) stability: This is an obvious prerequisite but we have seen quit a few project servers fail completely (uber face plant) within minutes of a challenge start.

*Work Unit availability: Like above, a challenge won't go far without work. SETI@Home suffers from this as do other projects with limited wu generation.

*Runtime: It helps volunteers if they know upfront what the estimated wu runtime, or est. range of runtimes for wu of varying sizes, will be necessary. Some projects will generate wu specifically geared toward a challenge event to make the event more interesting.

Scheduling Conflicts: Running a challenge that conflicts or overlaps with one or more other challenges can make the challenge a non-event. More evident when the other event is from a popular project. Especially from one that offers higher credits per hour.

Event Duration: Simply, if its a long running challenge, most volunteers won't bother participating.

Challenge Credits: Although this does not happen often, a project may increase credits awarded with the intent of drawing greater interest in the challenge.

OK, that's enough from me... :P Thank you again for taking to time to stop by and visit. I hope your 1st event proves to be a hit.

Best wishes,
F$

litehouse43
12-21-12, 12:01 PM
^^ Wow excellent information, thanks Fire$torm. Hopefully we will get something up and running in the beginning of 2013.

Fire$torm
12-21-12, 05:28 PM
Great. Looking forward to seeing another project on the network.

Best wishes to you and your team,
F$

Draconian
12-22-12, 03:51 AM
I think it is important for the project to realize, that regardless of their science - there are a lot of people crunching that are after numbers. If that person crunches for 3 hours on a task - and is awarded, say, 50 credit - they will leave in a heartbeat for a place with a ton more credit.
I don't need to be an expert on what the project is doing - general knowledge is enough (I'm bright enough). I need to be an expert on system building and optimization - and there, I'm fine. I work in I.T., have been for nearly 30 years - that's my field - that is my expertise. If you want to talk to me about protein folding and the intricacies of it - well - don't need to know - I'll take the scientists word and crunch.
Figure that a lot of these scientists behind the projects couldn't build a proper cruncher or fix their communications infrastructure to save their collective butts - and that is what I do. We all have a niche - what we are good at - what interests us. Years ago - I was trying to learn everything - name the programming language, the networking protocol, the means of transmission (sat, fiber), the muliplexing (DWDM is AWESOME!) - whatever - and I found that if you try to learn everything at once - you learn nothing. I don't want to be an expert in protein folding or figuring the optimum doughnut to doughnut hole ratio - there are only 24 hours in my day.

Credit for work done and availability are important - when the system is ready for work, it should have it. Optimized applications are a plus - nice to know that the system is doing what it can and not crunching on crap coding. Competition is a great thing - nice to contribute to a team effort. I guess the largest thing - and one that I think all of us could get behind - is if/when a project has a true and valid scientific achievement targeted - something that will UNDOUBTEDLY say, cure cancer, or aids, or whatever when it is completed - a REAL objective. I'd love to be a part of that. Finding the 34 millionth prime number - well...other than crunching and bragging rights - where is the BENEFIT? Does knowing the 34 millionth prime number do...anything...at all? I only use that as an example.

If there were a project with a REAL and definite accomplishment at the end - this WILL cure X - this WILL prove X (and it's important to prove it - not just scientists justifying their funding) - I'll crunch it for all it's worth.
You want real controversy - 2 projects - 1 to prove that God exists - the other to prove that he/she doesn't......OUCH!