> So right GG,
>
> Thats the beauty of BOINC. All projects will have their bad days (weeks,
> months *hehehe*), but Resource sharing is the key.
>
> Personally I will ALWAYS have SETI crunching away at some level... but now
> instead of worrying (can you believe how we WORRY when our CPU is sitting
> idle) when SETI is down now I rest assured that other work is being done.
>
> Personally right now I have all projects across the board equal. Doing some
> work for all worthy projects. Giving members personal preference to divide to
> what projects is one of the GREAT GREAT things about BOINC... but I always
> recommend that members run 2 projects ... just in case.
>
> Zain
> ---
>
>
Only thing is I wish you could devote 100% to a project and if it goes down it goes to the next project untill project A comes up. Where with resourse sharing you can miss some deadlines and einstein for example end up getting no credit even though your computer toiled away for 4 hours on a unit. So its hard to find a balance to use other projects or run around and attach to something else if your main project goes down.
It seems to me that we crunchers who have our "Connect ever X days" set low are not having troubles with deadlines (with one exception that I can think of, but Bruce seems to think it may be a bug).
I know this may not be very suitable for dial-up users, but for us with permanent connections I am advocating a reduction in the number of WU's being stored by an individual host.
> > So right GG,
> >
> > Thats the beauty of BOINC. All projects will have their bad days
> (weeks,
> > months *hehehe*), but Resource sharing is the key.
> >
> > Personally I will ALWAYS have SETI crunching away at some level... but
> now
> > instead of worrying (can you believe how we WORRY when our CPU is
> sitting
> > idle) when SETI is down now I rest assured that other work is being
> done.
> >
> > Personally right now I have all projects across the board equal. Doing
> some
> > work for all worthy projects. Giving members personal preference to
> divide to
> > what projects is one of the GREAT GREAT things about BOINC... but I
> always
> > recommend that members run 2 projects ... just in case.
> >
> > Zain
> > ---
>
> Only thing is I wish you could devote 100% to a project and if it goes down it
> goes to the next project untill project A comes up. Where with resourse
> sharing you can miss some deadlines and einstein for example end up getting no
> credit even though your computer toiled away for 4 hours on a unit. So its
> hard to find a balance to use other projects or run around and attach to
> something else if your main project goes down.
If BOINC was made smart enough to not download WU’s from projects that wouldn’t get CPU time in the next couple of days, we could set it up to crunch WU’s serial instead of parallel like today. Then deadlines wouldn’t be a problem, even if we set resource chare to project A=10000 B=1.
Then you're really interested in a subject, there is no way to avoid it. You have to read the Manual.
> All; Thanks for your comments. This is why I abandoned ship @ Berkeley. The
> constant denial of service from the project turned my little farm into an
> expensive collection of boat anchors. Sure, I could email and surf, but work?
> I did't make a serious investment in computer hardware to have it sit idle on
> the beach and contribute nothing to tommorrow. My crunching farm has been
> seriously happy with E@H since 22 January. My congratulations to all the
> grunts that keep the servers delivering the mail. Thanks again!
What I really don't understand is all of the angst.
First, "denial of service" is just the wrong term. They may not have been able to provide work, but you were (and still are) free to crunch other projects.
Second, why did you invest money? I still can't figure out why people would buy computers to build farms when we've got more computing than work.
But, it's the whole "kick 'em while they're down" syndrome. Everyone here who is praising Einstein and trashing SETI is able to crunch Einstein because of BOINC, and BOINC was produced by the same folks who developed SETI Classic.
> If BOINC was made smart enough to not download WU’s from projects that
> wouldn’t get CPU time in the next couple of days, we could set it up to
> crunch WU’s serial instead of parallel like today. Then deadlines wouldn’t
> be a problem, even if we set resource chare to project A=10000 B=1.
Actually, it might be. The server knows how long it takes you to return work units based on past history, and it certainly seems that it is adjusting how much work it offers based on how fast stuff comes back.
... and if you are crunching more projects, keeping "days between connections" low seems like a good thing.
> What I really don't understand is all of the angst.
>
> First, "denial of service" is just the wrong term. They may not have been
> able to provide work, but you were (and still are) free to crunch other
> projects.
I agree with you here. "denial of service" gives the conotation of a deliberate act intended to cause problems to others. SETI has not shut their project down deliberately to cause problems. Granted, they do have to take it down for maintainence on occasion, but that is usually announced in advance, or work arounds made to lessen the impact.
>
> Second, why did you invest money? I still can't figure out why people would
> buy computers to build farms when we've got more computing than work.
I don't understand why anyone would buy a huge V-10 Pick-up truck just to drive back and forth from home to the office, when they could buy a gas/electric car and get a million times more distance on their fuel economy.... but they do. It is a personal choice. I don't know exactly why these projects intrique me so much, but they do. I am one of them who would spend more money on a bigger computer farm if I had the extra money.
>
> But, it's the whole "kick 'em while they're down" syndrome. Everyone here who
> is praising Einstein and trashing SETI is able to crunch Einstein because of
> BOINC, and BOINC was produced by the same folks who developed SETI Classic.
Again, I have to agree with you here. Too many people make it sound as if SETI is being poorly managed or directed or that they are doing things intentionally to upset the "population". Granted, I know that there are plenty of improvements that the SETI team could make, but remember this, resources are limited. In ANY project or adventure that you or I embark upon, resources are limited. If the SETI team more just one more person, they might put him on as PR officer. If the SETI team had 25 hours in a day they might use that extra hour to completely log their daily events for us to read. If they had one more dollar, they might be able to buy more servers or UPS's, or more bandwidth.
Hmmm.... I guess I started ranting. Ok I will now relinquish my soap box. :)
Jim; You have an excellent perspective. My thanks for your comments. I only mean to say that I am committed to any project that can deliver the same support that I am willing to give. I hump spare power supplies and adjust vcore and fsb to get the max out of my hardware investment. I pay the utility bill as a matter of course. I upgrade or rebuild my crunchers when a deal comes along. I have little faith in the 400 celeron left on overnight. I applaud all those kind folks who volunteer for E@H. Please forword my welcome to our new participants. I simply enjoy contributing to a computing project that gives me 110% support.
> I agree with you here. "denial of service" gives the conotation of a
> deliberate act intended to cause problems to others. SETI has not shut their
> project down deliberately to cause problems. Granted, they do have to take it
> down for maintainence on occasion, but that is usually announced in advance,
> or work arounds made to lessen the impact.
More important, "denial of service" means more than just not delivering content. It means trying to completely tie up resources (bandwidth, sockets, file handles, whatever) to the point that the machine can't do anything.
My fascination is with the overall design of BOINC, and how people react to BOINC especially if they came from SETI Classic.
For Classic, if the servers were down, you finished a WU and you waited. Yeah, there were add-ons, but those were added later, not designed in.
For BOINC, the overall task is spread across servers so that one machine can theoretically be down and the others will pick up the slack. Work is cached in the client so if the project is down for an hour of maintenance, everyone keeps going.
I'm just now out of SETI work. I've got lots of Einstein and LHC. When SETI is back, I'll have SETI to do again.
From a technology standpoint, BOINC works pretty well. I'm sure it will be refined with time, but the basics are there. If you basically find some parameters and leave it alone, it'll handle outages like SETI is having, and it'll handle when LHC runs out of work.
By the way, I've learned about UPSes the hard way. When the batteries are new, they work great. When they get a little older you still have lots of time, and when they finally get tired, you think you have an hour on batteries and they're gone in 30 seconds. It's best to run them flat every few months and time the results (or run software that tests them periodically).
I understand and share your dedication to these projects. Although I think you may really get down into the trenches more by trying to tweak your machines. That is something that I don't know anything about and may be just a bit afraid to try. I don't want to lose what I do have because I cannot afford to replace anything when it goes bad. :) I also want you to know that I mean for none of my comments to be taken personally. My comments and observations were based upon many posts from many people over multiple projects. Well, I guess the comment about DoS was a little more specific to your post, but it still was not intended to belittle you.
So, with that said, let me ask you this... From what you've said I get the impression that you feel like SETI is not giving their full support to their project or that maybe you are feeling like you are more dedicated to their project than they seem to be. Am I close in my interpretation or way off base?
Jim
> Jim; You have an excellent perspective. My thanks for your comments. I only
> mean to say that I am committed to any project that can deliver the same
> support that I am willing to give. I hump spare power supplies and adjust
> vcore and fsb to get the max out of my hardware investment. I pay the utility
> bill as a matter of course. I upgrade or rebuild my crunchers when a deal
> comes along. I have little faith in the 400 celeron left on overnight. I
> applaud all those kind folks who volunteer for E@H. Please forword my welcome
> to our new participants. I simply enjoy contributing to a computing project
> that gives me 110% support.
>
> So right GG, > > Thats the
)
> So right GG,
>
> Thats the beauty of BOINC. All projects will have their bad days (weeks,
> months *hehehe*), but Resource sharing is the key.
>
> Personally I will ALWAYS have SETI crunching away at some level... but now
> instead of worrying (can you believe how we WORRY when our CPU is sitting
> idle) when SETI is down now I rest assured that other work is being done.
>
> Personally right now I have all projects across the board equal. Doing some
> work for all worthy projects. Giving members personal preference to divide to
> what projects is one of the GREAT GREAT things about BOINC... but I always
> recommend that members run 2 projects ... just in case.
>
> Zain
> ---
>
>
Only thing is I wish you could devote 100% to a project and if it goes down it goes to the next project untill project A comes up. Where with resourse sharing you can miss some deadlines and einstein for example end up getting no credit even though your computer toiled away for 4 hours on a unit. So its hard to find a balance to use other projects or run around and attach to something else if your main project goes down.
It seems to me that we
)
It seems to me that we crunchers who have our "Connect ever X days" set low are not having troubles with deadlines (with one exception that I can think of, but Bruce seems to think it may be a bug).
I know this may not be very suitable for dial-up users, but for us with permanent connections I am advocating a reduction in the number of WU's being stored by an individual host.
Jim
Jim
I run about 4 WU's for each
)
I run about 4 WU's for each project i'm attached to.
> > So right GG, > > > >
)
> > So right GG,
> >
> > Thats the beauty of BOINC. All projects will have their bad days
> (weeks,
> > months *hehehe*), but Resource sharing is the key.
> >
> > Personally I will ALWAYS have SETI crunching away at some level... but
> now
> > instead of worrying (can you believe how we WORRY when our CPU is
> sitting
> > idle) when SETI is down now I rest assured that other work is being
> done.
> >
> > Personally right now I have all projects across the board equal. Doing
> some
> > work for all worthy projects. Giving members personal preference to
> divide to
> > what projects is one of the GREAT GREAT things about BOINC... but I
> always
> > recommend that members run 2 projects ... just in case.
> >
> > Zain
> > ---
>
> Only thing is I wish you could devote 100% to a project and if it goes down it
> goes to the next project untill project A comes up. Where with resourse
> sharing you can miss some deadlines and einstein for example end up getting no
> credit even though your computer toiled away for 4 hours on a unit. So its
> hard to find a balance to use other projects or run around and attach to
> something else if your main project goes down.
If BOINC was made smart enough to not download WU’s from projects that wouldn’t get CPU time in the next couple of days, we could set it up to crunch WU’s serial instead of parallel like today. Then deadlines wouldn’t be a problem, even if we set resource chare to project A=10000 B=1.
Then you're really interested in a subject, there is no way to avoid it. You have to read the Manual.
> All; Thanks for your
)
> All; Thanks for your comments. This is why I abandoned ship @ Berkeley. The
> constant denial of service from the project turned my little farm into an
> expensive collection of boat anchors. Sure, I could email and surf, but work?
> I did't make a serious investment in computer hardware to have it sit idle on
> the beach and contribute nothing to tommorrow. My crunching farm has been
> seriously happy with E@H since 22 January. My congratulations to all the
> grunts that keep the servers delivering the mail. Thanks again!
What I really don't understand is all of the angst.
First, "denial of service" is just the wrong term. They may not have been able to provide work, but you were (and still are) free to crunch other projects.
Second, why did you invest money? I still can't figure out why people would buy computers to build farms when we've got more computing than work.
But, it's the whole "kick 'em while they're down" syndrome. Everyone here who is praising Einstein and trashing SETI is able to crunch Einstein because of BOINC, and BOINC was produced by the same folks who developed SETI Classic.
> If BOINC was made smart
)
> If BOINC was made smart enough to not download WU’s from projects that
> wouldn’t get CPU time in the next couple of days, we could set it up to
> crunch WU’s serial instead of parallel like today. Then deadlines wouldn’t
> be a problem, even if we set resource chare to project A=10000 B=1.
Actually, it might be. The server knows how long it takes you to return work units based on past history, and it certainly seems that it is adjusting how much work it offers based on how fast stuff comes back.
... and if you are crunching more projects, keeping "days between connections" low seems like a good thing.
> What I really don't
)
> What I really don't understand is all of the angst.
>
> First, "denial of service" is just the wrong term. They may not have been
> able to provide work, but you were (and still are) free to crunch other
> projects.
I agree with you here. "denial of service" gives the conotation of a deliberate act intended to cause problems to others. SETI has not shut their project down deliberately to cause problems. Granted, they do have to take it down for maintainence on occasion, but that is usually announced in advance, or work arounds made to lessen the impact.
>
> Second, why did you invest money? I still can't figure out why people would
> buy computers to build farms when we've got more computing than work.
I don't understand why anyone would buy a huge V-10 Pick-up truck just to drive back and forth from home to the office, when they could buy a gas/electric car and get a million times more distance on their fuel economy.... but they do. It is a personal choice. I don't know exactly why these projects intrique me so much, but they do. I am one of them who would spend more money on a bigger computer farm if I had the extra money.
>
> But, it's the whole "kick 'em while they're down" syndrome. Everyone here who
> is praising Einstein and trashing SETI is able to crunch Einstein because of
> BOINC, and BOINC was produced by the same folks who developed SETI Classic.
Again, I have to agree with you here. Too many people make it sound as if SETI is being poorly managed or directed or that they are doing things intentionally to upset the "population". Granted, I know that there are plenty of improvements that the SETI team could make, but remember this, resources are limited. In ANY project or adventure that you or I embark upon, resources are limited. If the SETI team more just one more person, they might put him on as PR officer. If the SETI team had 25 hours in a day they might use that extra hour to completely log their daily events for us to read. If they had one more dollar, they might be able to buy more servers or UPS's, or more bandwidth.
Hmmm.... I guess I started ranting. Ok I will now relinquish my soap box. :)
Jim
>
Jim
Jim; You have an excellent
)
Jim; You have an excellent perspective. My thanks for your comments. I only mean to say that I am committed to any project that can deliver the same support that I am willing to give. I hump spare power supplies and adjust vcore and fsb to get the max out of my hardware investment. I pay the utility bill as a matter of course. I upgrade or rebuild my crunchers when a deal comes along. I have little faith in the 400 celeron left on overnight. I applaud all those kind folks who volunteer for E@H. Please forword my welcome to our new participants. I simply enjoy contributing to a computing project that gives me 110% support.
> I agree with you here.
)
> I agree with you here. "denial of service" gives the conotation of a
> deliberate act intended to cause problems to others. SETI has not shut their
> project down deliberately to cause problems. Granted, they do have to take it
> down for maintainence on occasion, but that is usually announced in advance,
> or work arounds made to lessen the impact.
More important, "denial of service" means more than just not delivering content. It means trying to completely tie up resources (bandwidth, sockets, file handles, whatever) to the point that the machine can't do anything.
My fascination is with the overall design of BOINC, and how people react to BOINC especially if they came from SETI Classic.
For Classic, if the servers were down, you finished a WU and you waited. Yeah, there were add-ons, but those were added later, not designed in.
For BOINC, the overall task is spread across servers so that one machine can theoretically be down and the others will pick up the slack. Work is cached in the client so if the project is down for an hour of maintenance, everyone keeps going.
I'm just now out of SETI work. I've got lots of Einstein and LHC. When SETI is back, I'll have SETI to do again.
From a technology standpoint, BOINC works pretty well. I'm sure it will be refined with time, but the basics are there. If you basically find some parameters and leave it alone, it'll handle outages like SETI is having, and it'll handle when LHC runs out of work.
By the way, I've learned about UPSes the hard way. When the batteries are new, they work great. When they get a little older you still have lots of time, and when they finally get tired, you think you have an hour on batteries and they're gone in 30 seconds. It's best to run them flat every few months and time the results (or run software that tests them periodically).
Tweakster, I understand
)
Tweakster,
I understand and share your dedication to these projects. Although I think you may really get down into the trenches more by trying to tweak your machines. That is something that I don't know anything about and may be just a bit afraid to try. I don't want to lose what I do have because I cannot afford to replace anything when it goes bad. :) I also want you to know that I mean for none of my comments to be taken personally. My comments and observations were based upon many posts from many people over multiple projects. Well, I guess the comment about DoS was a little more specific to your post, but it still was not intended to belittle you.
So, with that said, let me ask you this... From what you've said I get the impression that you feel like SETI is not giving their full support to their project or that maybe you are feeling like you are more dedicated to their project than they seem to be. Am I close in my interpretation or way off base?
Jim
> Jim; You have an excellent perspective. My thanks for your comments. I only
> mean to say that I am committed to any project that can deliver the same
> support that I am willing to give. I hump spare power supplies and adjust
> vcore and fsb to get the max out of my hardware investment. I pay the utility
> bill as a matter of course. I upgrade or rebuild my crunchers when a deal
> comes along. I have little faith in the 400 celeron left on overnight. I
> applaud all those kind folks who volunteer for E@H. Please forword my welcome
> to our new participants. I simply enjoy contributing to a computing project
> that gives me 110% support.
>
Jim