TGIF!

[altivo]

Boy am I glad that week is over. Things have got to slow down or I'm gonna fall off the merry-go-round.

Monday is off work, but still working as the gallery show is judged that day. I have two woven towels to get hemmed this weekend because they're already promised for the show. I also have a saddle blanket to clean. It's just dusty because it's been hanging on the wall over my desk for five years. They asked for prize winning items from the last ten years for a retrospective display, and I was specifically invited to bring that item back.

Clear, cool, and dry, the way autumn should be. From the projections I saw last night, much of Canada was treated to an aurora display, but nothing was visible here. I got up at 5 am, with the intention of mucking stalls before work, but didn't get that far. So I had to do them after I got home from work at 6 pm. It's not very strenuous, but does take some time. Since 6 pm is normally feeding time, the horses and sheep were a bit miffed at having to wait 45 minutes or more before getting their suppers. Everyone is happy now though. Finished up after sunset and got back into the house at 7:30, by which time the dogs were nagging me for their own dinners. No word from Gary tonight. I suspect wherever he is up in Wisconsin now, he's found his cell phone is out of range.

I added two more machines to my BOINC flock today, so I have six running all together. SETI@Home and FightAIDS@Home are starting to rack up points already, and my scores are going to the FurSETI and Us Furries teams respectively. Einstein@Home is grinding away, but the work unit is so large that it won't be completed for two or three days yet. The only one I've had no success with at all is the ClimatePrediction.net. The work unit they gave me was gigantic, projected to take over 5000 hours of CPU time to complete. But when it tried to load and execute, it bombed. Ran out of some resource, I suspect, either memory or disk space. I've detached it rather than let it keep downloading result units and fail at them, after it rejected two in a row the same way. Maybe I just can't run their stuff.

Scores as of tonight:

SETI@Home shows 262 credits granted and 57 more pending
FightAIDS@Home shows 339 points awarded

I'm not sure how "points" compare to "credits" actually. Even with those paltry numbers, I've jumped up in the ranks and am no longer near the bottom of the lists. One more work unit should finish for each before midnight tonight. I may not be awake when it happens, though.

Comments

Hi.

[goldenstallion.livejournal.com]

Whinnyhi.

I know exactly what you need to do. Come comfort and visit me no hooves barred. Oh well so I will continue to struggle with my fursuit in hopes it might be finished in time. Doubt it very seriously.

Steed

Re: Hi.

[altivo.livejournal.com]

I'd love to do that, but I'm not sure when we could fit it in before the con. The gallery show in Woodstock has me tied down on weekends in October, but I'll look more closely at the calendar.

You're far enough along that you can always do it as a partial suit this time. Hands, hooves, and that fine head you've done, combined with suitable pirate clothes or whatever costume you choose. :)

[heavens-steed.livejournal.com]

Yay for merry-go-rounds! ^^

What is the point of those programs? I've seen that SETI@Home before but I don't understand what they do. What do the points mean?

[altivo.livejournal.com]

SETI is the Search for Extraterrestrial Intelligence. The research project uses radio data from the very large radiotelescope dish at Arecibo in Puerto Rico, and does computer analysis of all the radio signals received on a range of frequencies that theoretically would be used for communication by intelligent beings more or less like ourselves. The frequencies are chosen for several different reasons.

Of course we don't know exactly what sort of information would be contained in an alien radio signal. It might not be directed at us, but rather used for their own purposes, to communicate with satellites or spacecraft, or with other nearby planets. But a signal that contains intelligible information is bound to be modulated in some pattern that can be distinguished from background noise, and that's what the computers look for.

The project was originally run on supercomputers (the famous Cray machines, for instance) because it involves a tremendous amount of math. When they began to run out of money to rent time on those machines, the idea came up of developing a distributed computing network and letting volunteers donate computer time. That's when SETI@Home was born. A program similar to a screen saver, that runs only on unused CPU time, is installed on your PC. It downloads chunks of data over the internet, processes them in the background, and returns the results to the project's central site at the University of California in Berkeley. Check their web site for more details.

Though SETI@Home hasn't yet found proof of alien intelligence, it has proved that distributed computing of huge mathematical problems by volunteers can be a successful concept. Dozens of other projects have now sprung up that use a similar technique. Some are pure science, like mathematical searches for huge prime numbers, while others are related to physics or astronomy, like Einstein@Home, which hunts through radio telescope data for as yet undiscovered pulsars. Fairly new are the projects that study DNA and protein folding by analyzing molecular structures and modifying them a few atoms at a time. Those may contribute to the development of new drugs and vaccines. Other projects study epidemiology, climatology, and even social interaction through mathematical models.

The scores are just for volunteer recognition. Each unit of work done on your computer earns you a number of points defined by the amount of resources it used and possibly any useful results returned. It's sort of like handing out certificates of merit. You can join a team and compete with other groups to see who gets the most points the fastest, for instance. That part is entirely optional, of course, but does provide some added incentive for many people.

Since corelog has been an avid participant for a while, I'm surprised he hasn't explained it to you. He probably knows more about it than I do, since I've only just begun to volunteer my extra computer time. I'm sharing time out to three projects: SETI@Home, Einstein@Home, and FightAIDS@Home (which is part of the larger project, World Community Grid, largely concerned with health and medical issues like AIDS and Dengue Fever.)

[corelog.livejournal.com]

I think I didn't tell Finrod only because he wasn't online at any point during my "Install BOINC!!!" drives. Otherwise, I definitely would have included him. :)

[heavens-steed.livejournal.com]

I already knew what SETI was, I just didn't know what this program was all about and what it did. Thank you for the detailed explanations, although now I feel another degree more stupid.

[altivo.livejournal.com]

Why? Not knowing what something is does not make you stupid. Stupid is not being able to understand when it's explained or described. You obviously understood the explanations, at least the basis of it, so I'd say you're NOT stupid.

If you feel stupid because you think you couldn't run this stuff, you're just wrong. You could if you wanted to. They make it very easy to set up. Then you just pick one or more projects and let them go. They take care of themselves, downloading new data when necessary and uploading their results when ready. The programs even deal well with an internet connection that goes up and down as mine does. When the link is down, they just wait and check again every so often until it comes back up.

I tend to watch and see what they're doing and how fast, and shift tasks around, but that's not necessary. I'm just curious about it. All the statistic keeping and scoring happens automatically.

[corelog.livejournal.com]

Tivo already covered it fairly well, actually. Basically, we crunch data for science. SETI uses our computers to search for radio signals in the background noise of the cosmos. World Community Grid uses our computers to study drug molecule interactions between viruses (such as dengue, malaria, and HIV), or to analyze how human proteomes fold up to make various protein structures in the body which can then be used to better design drugs to cure various diseases. Einstein uses our computers to analyze data from the Laser Interferometry Gravitational Observatory to see if we can detect the faint vibrations from gravitational waves radiated by spinning pulsars. Climateprediction.net uses our computers to run climate models on the globe, with every participant being given slightly different models to run for 160 years of model time, and see how they change based on those tiny differences.

I assign my computers to projects based on how strong my computers are. Generally, I'll take any Pentium II or higher, and every machine is given at least 256MB of RAM (CPDN requires 512MB in order to run the more advanced coupled model, instead of the simplistic slab model).

I could go on and on for a couple hours about BOINC. :) I definitely recommend installing it IF and only if you leave your computer on a lot of the time. This is why BOINC is usually installed on desktops more than laptops, especially laptops that travel a lot. But if, like Syrus, you leave your laptop connected and doing mostly nothing while you're at work, BOINC is a great use of that spare time. Of course, the dragon still won't install it, because he's too lazy. Feels that three minutes would be too much work. Lazy and wasteful, as I've told him.

[altivo.livejournal.com]

Can you actually get CPDN to run on those machines? Their work units kept bouncing off both my Pentium 4 class machines. Each has a gigabyte of RAM and scads of disk space, but as soon as their unit would start it would abort, complaining that it couldn't get enough shared memory. I've detached them completely for now as there seems to be little point in wasting bandwidth and CPU on just downloading their stuff only to have it fail. The other three projects I'm on are all chugging along, though Einstein has yet to turn up actual results.

[corelog.livejournal.com]

I don't run CPDN except on machines running upwards of 2GHz in speed. Hence, I have two machines (one here and my work computer) that run CPDN. The one here uses Linux, the one at work uses Windows using Hyper-Threading to do two CPDN units at once.

Unfortunately, CPDN is also extremely sensitive to minor bugs and glitches in the OS and hardware. It's not uncommon for machines to have two workunits fail out of every five, sometimes one right after the other. A higher failure rate usually means your system just isn't stable enough for CPDN, though there are some tips that can help stabilize the system. The Unofficial BOINC Wiki has lots of pages relating to CPDN, including this one: http://www.boinc-wiki.info/index.php?title=How-To_Test_Machine_Stability

[altivo.livejournal.com]

Well, I read the article but frankly, I'm not impressed with the idea of using CPDN's own software as a measure of "stability". I suspect it's partly the software that's to blame. They are asking people to donate valuable processing time to them for free, without even a tax deduction. Then they say "If our mysterious software won't run on your machine, it's the machine that's at fault." Sorry, no. I've heard that before, most notably from MicroSloth themselves, and they weren't right either. Eventually they admitted I was right and they were wrong, but without any apology. Since I don't run Windows, the suggestions given in that article are mostly irrelevant anyway.

I dropped CPDN after two failures in a row. Especially since their workunits are so large and long-running, I'm not going to have one eating up my time for weeks and then suddenly crash for no reason. Better the time go to a project that can make effective use of it. ;p

[heavens-steed.livejournal.com]

I already knew what SETI was, but I didn't know what the program was or its purpose. I asked because I saw SETI@Home running on Jim's computer. He also has one that does something with proteins. Right now I tend to leave my laptop running more often than I normally do since I don't have to be anywhere for a long period of time having no job or school. But usually when I go to work, my machine goes off.

Thanks for the explanation but now I am reminded how stupid I am...

Altivo Horsey?

[(Anonymous)]

Altivo Horsey !! :)

Re: Altivo Horsey?

[altivo.livejournal.com]

???

Perrier?

Re: Altivo Horsey?

[(Anonymous)]

Yes :)

Re: Altivo Horsey?

[altivo.livejournal.com]

And even terser than usual. ;p Hiya.

coasting...

[gabrielhorse.livejournal.com]

Things have got to slow down or I'm gonna fall off the merry-go-round.

I thought the same thing before I moved from Pensacola... now I almost feel like I'm being lazy XD

Re: coasting...

[altivo.livejournal.com]

Heh. "Lazy" is definitely not an adjective I'd apply to you.

Re: coasting...

[(Anonymous)]

Heh... yeah I know, I'm just saying by comparision that how i feel in my head :P I'm actually always doing something, but it seems lees strenuous or challenging compared to what I was doing a month ago... but I guess that's to be expected. I'd probably find mucking out stalls an easygoing time of it myself ;)

Re: coasting...

[gabrielhorse.livejournal.com]

Heh... yeah, well... it's how i felt a the time. I've actually stayed what most people would call busy, it just seems bland compared to what I've been through :P

[corelog.livejournal.com]

Points at WCG don't directly translate to BOINC credits, unfortunately. Just by looking at my own totals, it seems to run at something like 325,000 WCG points = 50,000 BOINC credits. Or something around that conversion rate. Most of my work units for WCG take about 10 hours of compute time and yield around 40 BOINC credits.

CPDN is the highest-yielding BOINC task as far as credits go. This is offset by the immense size and length of their workunits. :P There's a regular discussion about this on the BAM! shoutbox and forums. Speaking of BAM!, I've tried looking you up on it, but I can't seem to find you! Care to give me a direct link to your stats? :) I'd like to watch!

[altivo.livejournal.com]

Bam! statistics here.

I am not impressed with Bam!'s reporting though. They lag reality by 24 hours or more and I see currently they are mixed up about teams. They seem to want to attach me to whichever team the latest number was reported for, rather than recognizing that I report for two different teams depending on the project. I have a bunch of unreported credit right now, partly because SETI@Home has been rejecting reports this morning and partly because WCG is slow in approving stuff.

[corelog.livejournal.com]

Thanks for the link. :) BAM!'s statistics do lag a little, mainly because of the structure. They pull data from the projects' XML feeds and perform a daily update with that. In the case of CPDN in particular, any work I report on one day won't be reported until the next day, because CPDN only pushes new stats to their XML feed once every 24 hours. And depending on whether CPDN manages to flush all the stats to XML before BAM!'s daily update, I occasionally have to wait for a second day before the credits show up in my total.

[cabcat.livejournal.com]

Hehe, we don't really have much of an Autumn here as a lot of trees are Evergreen. Its really warming up nicely and all the flowers are out. Still with severe water restrictions in place I hope the summer is a bit wetter.

Sometimes I'll get World Community grid running at once on three computers *chuckles* Mainly in winter to help heat the house up.

[altivo.livejournal.com]

I read somewhere that Australia may be suffering drought already due to global warming. Is summer usually your wetter season, or would that be winter? I know climate varies a lot in a place as large as Australia, but you're near the coast, right? Seems as if you ought to get a fair amount of rain unless the wind always blows offshore.

I've noticed the warming effect of BOINC even with just two machines running it. I knew the Alpha would add heat to the room, which is fine with winter coming on. I can tell I'm going to have to clean heatsinks more often on the Intel processors, because their fans are running harder.

[cabcat.livejournal.com]

Oh we've been having a drought for a decade in some places. Summer is usually our rainy season but that hasn't happened for ages now.

*hands you a can of compressed air* Computer maintenance :) *imagines a plushie horse holding a can of compressed air between two paffers*

[altivo.livejournal.com]

Actually, I prefer to use a vacuum cleaner with one of those tiny adapters that lets you use itty-bitty brushes in there.