Hi,

During this kind of famine
a) would it be wise to let all GPU's process all kind of work? AMD GPUs have their caches much better filled. NVIDIA ones do not get arecibo vlars. I'd like to get some and I think that those running special 'sauce' would like too. If plain vanilla CUDA apps got vlars that would keep them busy. They would not be asking more for a long time.
b) to limit the servers to send only 25 work units to a host at a time (I can do 4*5*1.3 shorties and 4*5*.33 guppi vlars in 5 minutes). And the next wu's would be sent to a host only after some results have been sent back.

This is an unfinished idea and it may certainly contain some errors in calculations and I do not know anything about the inner workings of the servers.

Petri

---

To overcome Heisenbergs:
"You can't always get what you want / but if you try sometimes you just might find / you get what you need." -- Rolling Stones

That's because you have told the server you want AP work in your prefs but you have no app for it.

And even when I don't have that option selected I still get the messages in the Event log. Just getting it in the Manager Notices is new.

You can get a message for either MB or AP depending on your app_info.xml no matter what your app selections are.
The quirk in the messages tab/event log has been driving me nuts for years.
My AP only machines would get "you don't have a usable version of MB" message and my MB only machines would get the "you don't have a usable version of AP.
I found if I configured the app that I didn't want to run for hardware I didn't have then the message didn't seem to show up.
Like setting up the NV AP app on a machine that only ran CPU & ATI GPU MB work.

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[

Yes, the 1.18 app is about 500 seconds faster on average compared to 1.17. I too have noticed an increase in the number of tasks downloaded per request. Haven't seen 100 yet but I did see a bump from 20 or so previously to 40 or so the last time I asked for work. Einstein has no problems producing work on request for the most part. A lot more stable project compared to SETI. I am having serious thoughts about switching my priorities over to Einstein even though SETI was my first project and my main allegiance in the past. SETI is just taking too much time to personally manage lately.

. . I received 45 tasks on the unit with the 1050ti and double that on the unit with the 2 x 1060s. I didn't actually count them but it was enough to fill several screens making it about the 100 mark.

. . Einstein has some attractions but my heart is with SETI, even though the frustrations of constantly shrinking caches is straining my commitment. One thing I really, really like about E@H is they don't have creditscrew LOL. Each task gets the same number of points, and each task takes about the same time because they are configured that way. So the points are a consistent measurement of your contribution. That feels soooo right :)

. . But I sure will feel better when they guys solve this problem with the splitters/servers, I like knowing that when I am out the rigs will have enough work to justify them being turned on and using power. Otherwise they get to sleep.

Stephen

??

So, the mysterious thing that takes the BOINC server down... has just taken the BOINC server down again. Last time it was most of Friday, so I suspect that now it'll be most of the coming night. We'll see. (Can always blame the guy who wanted to make a new account. ;-))

It works for me. A bit slow, but it isn't down.

Edit: and now it isn't even slow. Works normally.

. . Braggart! I got full caches this morning for the first time in over 12 hours. I was getting ZERO work for that time.

Stephen

:(

Hi,

During this kind of famine
a) would it be wise to let all GPU's process all kind of work? AMD GPUs have their caches much better filled. NVIDIA ones do not get arecibo vlars. I'd like to get some and I think that those running special 'sauce' would like too. If plain vanilla CUDA apps got vlars that would keep them busy. They would not be asking more for a long time.
b) to limit the servers to send only 25 work units to a host at a time (I can do 4*5*1.3 shorties and 4*5*.33 guppi vlars in 5 minutes). And the next wu's would be sent to a host only after some results have been sent back.

This is an unfinished idea and it may certainly contain some errors in calculations and I do not know anything about the inner workings of the servers.

Petri

. . That might help those doing ONLY GPU crunching. But for me the Arecibo VLARS and the guppis are what keep my CPUs fed :)

. . The GPUs live on normal AR Arecibo work and the excess guppis that the CPUs can't digest ...

.. Having said that, when the situation gets too bad I will move whatever work is abundant (or at least relatively) from one queue to the other to keep all devices occupied. While slow, Arecibo VLARs will crunch happily under SoG, and I suspect under your Special Sauce though I have yet to find that out.

Stephen

:)

. . I received 45 tasks on the unit with the 1050ti and double that on the unit with the 2 x 1060s. I didn't actually count them but it was enough to fill several screens making it about the 100 mark.

. . Einstein has some attractions but my heart is with SETI, even though the frustrations of constantly shrinking caches is straining my commitment. One thing I really, really like about E@H is they don't have creditscrew LOL. Each task gets the same number of points, and each task takes about the same time because they are configured that way. So the points are a consistent measurement of your contribution. That feels soooo right :)

. . But I sure will feel better when they guys solve this problem with the splitters/servers, I like knowing that when I am out the rigs will have enough work to justify them being turned on and using power. Otherwise they get to sleep.

Stephen

??

I have the exact same sympathies. SETI will always hold a special place in my heart since it got me started with distributed computing back in Classic days. I agree completely on the credit mechanism at Einstein. Makes absolute sense and best shows the lunacy of CreditNew here at SETI.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

Well, just got home from work and am happy to see that the PFB splitters have been working their butts off trying to keep up with demand. Caches are full here, and there is just a little bit of RTS built up.
Hope they can keep it rolling.

Meow!

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

Well, just got home from work and am happy to see that the PFB splitters have been working their butts off trying to keep up with demand. Caches are full here, and there is just a little bit of RTS built up.
Hope they can keep it rolling.

Meow!

AMEN!

Stephen

.

The kitties are crunching a storm up with all the high protein kibbles..........
Meowmunchmunchmunch.

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

I have the exact same sympathies. SETI will always hold a special place in my heart since it got me started with distributed computing back in Classic days. I agree completely on the credit mechanism at Einstein. Makes absolute sense and best shows the lunacy of CreditNew here at SETI.

Agree 100%, but DA is God, until that changes, business as usual.

And some of us just take da lumps and carry on with the business at hand.
I am sure Dr. Anderson is aware of all of the complaints leveled in his general direction.
Just having reached the 830 MILLION credit marker, it appears to have been working OK for me.
I am not about to complain that it could be 850 million credits.

Meow.

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

Down to 7 GPU tasks on one computer now. Not getting any work when requested. Guess I will have to sit and play the project toggle game for 15 minutes now.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

Down to 7 GPU tasks on one computer now. Not getting any work when requested. Guess I will have to sit and play the project toggle game for 15 minutes now.

As I have said before, not sure what the problem is there.
And I am not sure Eric does either.
I am down about 15 short of a full load, so something is going on.
Could be that the current crop of data is just not splitting much GPU work.
In any case, no worries here.

I do know that I have not changed any settings on my accounts for months now./

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

Down to 7 GPU tasks on one computer now. Not getting any work when requested. Guess I will have to sit and play the project toggle game for 15 minutes now.

As I have said before, not sure what the problem is there.
And I am not sure Eric does either.
I am down about 15 short of a full load, so something is going on.
Could be that the current crop of data is just not splitting much GPU work.
In any case, no worries here.

I do know that I have not changed any settings on my accounts for months now./

I wish I knew the answer too. Part of why I am getting so frustrated with SETI. It is no longer a project that you can join and just let it run unattended as it was in the past. Needs constant monitoring to make sure it has work to crunch. I was at full complement on all 3 computers for over a day after they got the stuck splitters off that bad tape. That I think was the longest the computers had run without needing management since all the server upgrades. Woke up this morning and all computers were down to less than 25 GPU tasks on board. After I toggled the projects in Preferences, I am back to full strength again, but for how long is anybody's guess.

It would be nice if the SETI administrators would at least acknowledge that there is a problem and that they are working on it. Wishful thinking I know.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

While you feel you can't sit back and let BOINC do its thing there are those, like myself, who are quite happy to let BOINC do what it does - I can't remember the last time I did anything other than watch my top ten cruncher work its way through whatever tasks have been sent by the servers, on which ever processor that the server sent those tasks to. Indeed only one of my crunchers gets any more attention that that - my daily driver, and that just gets BOINC shutdown when it is doing money earning work.

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

While you feel you can't sit back and let BOINC do its thing there are those, like myself, who are quite happy to let BOINC do what it does - I can't remember the last time I did anything other than watch my top ten cruncher work its way through whatever tasks have been sent by the servers, on which ever processor that the server sent those tasks to. Indeed only one of my crunchers gets any more attention that that - my daily driver, and that just gets BOINC shutdown when it is doing money earning work.

If that were the case, I should just remove the project from my computers since if they are left unattended they will be out of work permanently in just a few days and won't get any work after that. That is the consequence of not constantly fiddling with the server settings and will be working on nothing more than my secondary projects which work quite well thank you completely unattended.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

While you feel you can't sit back and let BOINC do its thing there are those, like myself, who are quite happy to let BOINC do what it does - I can't remember the last time I did anything other than watch my top ten cruncher work its way through whatever tasks have been sent by the servers, on which ever processor that the server sent those tasks to. Indeed only one of my crunchers gets any more attention that that - my daily driver, and that just gets BOINC shutdown when it is doing money earning work.

If that were the case, I should just remove the project from my computers since if they are left unattended they will be out of work permanently in just a few days and won't get any work after that. That is the consequence of not constantly fiddling with the server settings and will be working on nothing more than my secondary projects which work quite well thank you completely unattended.

reading different topics on this forum shows that some people are relying on what Keith is doing to get work while others are not having to go through this like rob is saying.

It seems this particular problem is affecting some more than others... most of us are here because seti@home was our first project... and I think most of us are still here because it is a good community....

despite the issue with lack of work, the unfair credits system (even if you don't care about it, it is a problem), and whatever else is bothering you... if it wasn't for the community and what this project represents, why else would anyone stick to a project that returned 0 results since 1999?

---

My two fast computers burn through 750 GPU tasks and 75 CPU tasks a day. My slow computer does 500 GPU/50 CPU tasks a day. I'm not the only one with this problem. Grant is also greatly affected and I know there are others that have commented on this problem in this thread. We are the "random" hosts with the problem who have to constantly fiddle.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

It is worth remembering that the project has NEVER guaranteed us a "never ending" supply of tasks.
The advise, from the project, is to have at least one standby project, so that if one's main project runs out of tasks, for whatever reason. This can be achieved by setting your standby project priority to zero, thus it will only get tasks when required.

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

While you feel you can't sit back and let BOINC do its thing there are those, like myself, who are quite happy to let BOINC do what it does - I can't remember the last time I did anything other than watch my top ten cruncher work its way through whatever tasks have been sent by the servers, on which ever processor that the server sent those tasks to. Indeed only one of my crunchers gets any more attention that that - my daily driver, and that just gets BOINC shutdown when it is doing money earning work.

If that were the case, I should just remove the project from my computers since if they are left unattended they will be out of work permanently in just a few days and won't get any work after that. That is the consequence of not constantly fiddling with the server settings and will be working on nothing more than my secondary projects which work quite well thank you completely unattended.

reading different topics on this forum shows that some people are relying on what Keith is doing to get work while others are not having to go through this like rob is saying.

It seems this particular problem is affecting some more than others... most of us are here because seti@home was our first project... and I think most of us are still here because it is a good community....

despite the issue with lack of work, the unfair credits system (even if you don't care about it, it is a problem), and whatever else is bothering you... if it wasn't for the community and what this project represents, why else would anyone stick to a project that returned 0 results since 1999?

For the most part, BOINC does its thing fairly well ... other than some occasional issues with project priority scheduling. It's a pretty robust underlying architecture for distributed computing projects now. I don't think I have done anything to MilkyWay in at least a couple of years. It just downloads new work and applications without my intervention and just runs. Always have work. Same pretty much for Einstein. You just have to monitor any new campaigns that pop up occasionally and are automatically enabled in your project preferences which you might have to turn off if you don't want that work. It too doesn't seem to have any problems creating work for its clients barring the occasional server maintenance periods. Also, I like that they always announce scheduled server maintenance well in advance of the date and nobody is ever surprised of the temporary interruptions in delivering work. I wish SETI was run as well these days. Lately, SETI is always in crisis mode, running to put out fires constantly. I am having buyers remorse over my yearly contribution to the project. I'm wishing my money had gone to Einstein last year. I know that I am not the only one who has become disheartened with the project. I might be joining Zalster over at Einstein as my primary project.

I agree that this has a great support community and one of the reasons I remain so far. There is very little traffic in Number Crunching over at MilkyWay. Same for Einstein in Cruncher's Corner which has a bit more than MW. Some of that is because not much changes except for the occasional application update. Mainly there is little forum traffic because the projects JUST work and don't need much troubleshooting input from forum members.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

It is worth remembering that the project has NEVER guaranteed us a "never ending" supply of tasks.
The advise, from the project, is to have at least one standby project, so that if one's main project runs out of tasks, for whatever reason. This can be achieved by setting your standby project priority to zero, thus it will only get tasks when required.

Yes, I know and why I have Einstein and MilkyWay as my backup projects. I just might relegate SETI as my backup project and push Einstein as primary.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)