You know the drill.

---

huuh, where is all the work of astropulse gone, I see 4 splitter running
no work there since days


See Matt's last 'technical news' post....
I am not quite sure what is going on.......he seems to be saying that they want to put out more MB work, and less AP....

This seems like going backwards to me.

(For continued discussion in the new thread).

---

"Time is simply the mechanism that keeps everything from happening all at once."

huuh, where is all the work of astropulse gone, I see 4 splitter running
no work there since days


See Matt's last 'technical news' post....
I am not quite sure what is going on.......he seems to be saying that they want to put out more MB work, and less AP....

This seems like going backwards to me.

(For continued discussion in the new thread).

Joe Segur did an excellent analysis of the appropriate proportions of AP to MB (but I can't lay hands on the post ATM to point to it).

F.

---

Joe Segur did an excellent analysis of the appropriate proportions of AP to MB (but I can't lay hands on the post ATM to point to it).

Message 897878 in "More AP WU's please!" a week ago pretty much covers it.

In response to Message 900816:

Ned Ludd wrote:

The BOINC client and the BOINC servers have infinite patience, and left alone, work gets reported and new work downloaded.

The BOINC client has finite patience. If the upload, download or report hasn't happened after 14 days, it's completely aborted and deleted.

The BOINC servers, in the form of the database, wait only till the deadline on the work and then abort trying to send that work to you.

---

huuh, where is all the work of astropulse gone, I see 4 splitter running
no work there since days


See Matt's last 'technical news' post....
I am not quite sure what is going on.......he seems to be saying that they want to put out more MB work, and less AP....

This seems like going backwards to me.

(For continued discussion in the new thread).

I recall reading that they have reduced the amount of AP WU's being released so that the MB WU's can get caught up. Check the server status page. there are dozens of tapes not processed for MB but finished for AP. We are all going to have to suffer running the MB for a while

---

In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope

In response to Message 900816:

Ned Ludd wrote:

The BOINC client and the BOINC servers have infinite patience, and left alone, work gets reported and new work downloaded.

The BOINC client has finite patience. If the upload, download or report hasn't happened after 14 days, it's completely aborted and deleted.

The BOINC servers, in the form of the database, wait only till the deadline on the work and then abort trying to send that work to you.

Which, if we assume that an outage is a week or less, is practically infinite.

... and that was my point. There is a huge difference between "it didn't upload on the first try, PANIC!" and "I waited a day, and everything took care of itself."

SMTP works the same way, but the average E-Mail user is blissfully unaware of how much work mail servers do, and how often mail takes more than one attempt.

We both know that on average the SETI@Home servers can handle the load. There may be times when they're overloaded, but if we as SETIzens sit back and do nothing, the problems will either go away by themselves, or the folks at Berkeley will correct the failure and things will return to normal.

Waiting is....

---

Joe Segur did an excellent analysis of the appropriate proportions of AP to MB (but I can't lay hands on the post ATM to point to it).

F.

He did an excellent analysis.

My only thought is that each "tape" likely has "x" number of multibeam work units and "y" number of astropulse (i.e. there are a relatively constant number of seconds recorded, and a given number of seconds will always produce the same amount of work).

If the project tries to artificially adjust the amount of work to favor one or the other, it seems to me that if you favor AP, there will come a time when all you have is MB, and you can't maintain that favorable balance.

---

I believe he said that the mix would be 95% MB 4% AP and 1% rerun/retry of AP

---

In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope

I did not unsubscribe "Panic Mode On (14) Server problems" before it was closed. How can I get it off my subscription list?

---

I assume by opening the thread and the unsubscribing it

---

In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope

I had done that before and it still stayed. I just now tried subscribing again and then unscribing. It worked.

---

Joe Segur did an excellent analysis of the appropriate proportions of AP to MB (but I can't lay hands on the post ATM to point to it).

F.

He did an excellent analysis.

My only thought is that each "tape" likely has "x" number of multibeam work units and "y" number of astropulse (i.e. there are a relatively constant number of seconds recorded, and a given number of seconds will always produce the same amount of work).

If the project tries to artificially adjust the amount of work to favor one or the other, it seems to me that if you favor AP, there will come a time when all you have is MB, and you can't maintain that favorable balance.

I wish to offer a little apology for not quite understanding the dynamics of what is going on.......

I did not realize that the datasets included both AP and MB work on the same 'tape'....I thought they were separate entities.

If each incoming dataset includes a set amount of AP and MB work, I can see how it has come to pass that we have accumulated a large number of datasets (tapes) that have had all of the AP work issued and crunched, but still have the leftover MB work to be processed.

This is indeed a quantum shift........
Nobody wanted the AP work at first. Took too long.....then the Lunatics boyz stepped in and provided an app that would crunch them up smartly. And the kitties went wild..........

Suggestions......now that things have shifted a bit, change the default for new users to 'opt-in' for AP....the default should be MB.

And the other suggestion would be to raise the pay rate for optimized MB to match what optimized AP now offers.....
That would make it more palatable for AP crunchers like myself to back off on AP and crunch up some more of the current MB overload.

---

"Time is simply the mechanism that keeps everything from happening all at once."

Joe Segur did an excellent analysis of the appropriate proportions of AP to MB (but I can't lay hands on the post ATM to point to it).

F.

He did an excellent analysis.

My only thought is that each "tape" likely has "x" number of multibeam work units and "y" number of astropulse (i.e. there are a relatively constant number of seconds recorded, and a given number of seconds will always produce the same amount of work).

If the project tries to artificially adjust the amount of work to favor one or the other, it seems to me that if you favor AP, there will come a time when all you have is MB, and you can't maintain that favorable balance.

I wish to offer a little apology for not quite understanding the dynamics of what is going on.......

I did not realize that the datasets included both AP and MB work on the same 'tape'....I thought they were separate entities.

If each incoming dataset includes a set amount of AP and MB work, I can see how it has come to pass that we have accumulated a large number of datasets (tapes) that have had all of the AP work issued and crunched, but still have the leftover MB work to be processed.

This is indeed a quantum shift........
Nobody wanted the AP work at first. Took too long.....then the Lunatics boyz stepped in and provided an app that would crunch them up smartly. And the kitties went wild..........

Suggestions......now that things have shifted a bit, change the default for new users to 'opt-in' for AP....the default should be MB.

And the other suggestion would be to raise the pay rate for optimized MB to match what optimized AP now offers.....
That would make it more palatable for AP crunchers like myself to back off on AP and crunch up some more of the current MB overload.

Good luck on raising MB credits.

---

The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's

Good luck on raising MB credits.

It is only a suggestion based on the current state of affairs......

I am simply pointing out the obvious fact that some of us kitties are crunching opti AP because it pays better than opti MB....(and it supposedly is doing better science)...
If the opti MB pay rate was the same, I would not care what the kitties had the crunchers doing....

---

"Time is simply the mechanism that keeps everything from happening all at once."

Joe Segur did an excellent analysis of the appropriate proportions of AP to MB (but I can't lay hands on the post ATM to point to it).

F.

He did an excellent analysis.

My only thought is that each "tape" likely has "x" number of multibeam work units and "y" number of astropulse (i.e. there are a relatively constant number of seconds recorded, and a given number of seconds will always produce the same amount of work).

If the project tries to artificially adjust the amount of work to favor one or the other, it seems to me that if you favor AP, there will come a time when all you have is MB, and you can't maintain that favorable balance.

I wish to offer a little apology for not quite understanding the dynamics of what is going on.......

I did not realize that the datasets included both AP and MB work on the same 'tape'....I thought they were separate entities.

If each incoming dataset includes a set amount of AP and MB work, I can see how it has come to pass that we have accumulated a large number of datasets (tapes) that have had all of the AP work issued and crunched, but still have the leftover MB work to be processed.

This is indeed a quantum shift........
Nobody wanted the AP work at first. Took too long.....then the Lunatics boyz stepped in and provided an app that would crunch them up smartly. And the kitties went wild..........

Suggestions......now that things have shifted a bit, change the default for new users to 'opt-in' for AP....the default should be MB.

And the other suggestion would be to raise the pay rate for optimized MB to match what optimized AP now offers.....
That would make it more palatable for AP crunchers like myself to back off on AP and crunch up some more of the current MB overload.

I will agree with AP being an opt-in sort of deal, but after looking at one of the causes of all of my pending credits/tasks.. AP should not be allowed on hosts that have less than a 250 RAC. That requirement makes sure they're going to actually try to do some crunching. 90% of the wingmen I get have less than 5 tasks, are brand new rigs, or have RACs of less than 10..because of one/both of the previous two reasons.

---

Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)

I will agree with AP being an opt-in sort of deal, but after looking at one of the causes of all of my pending credits/tasks.. AP should not be allowed on hosts that have less than a 250 RAC. That requirement makes sure they're going to actually try to do some crunching. 90% of the wingmen I get have less than 5 tasks, are brand new rigs, or have RACs of less than 10..because of one/both of the previous two reasons.

Which bolsters my suggestion that MB credits be raised.......to get some of us AP enthusiasts to allow their rigs to crunch down the MB overload......

And for cryin' out loud, why make the newbies cut their teeth on AP when they don't know the difference anyway??? Start them out on MB, and let them opt into AP when they know and want the different work.............

---

"Time is simply the mechanism that keeps everything from happening all at once."

Looks like my videocard will be cooling down tonight...

---

me@rescam.org

I will agree with AP being an opt-in sort of deal, but after looking at one of the causes of all of my pending credits/tasks.. AP should not be allowed on hosts that have less than a 250 RAC. That requirement makes sure they're going to actually try to do some crunching. 90% of the wingmen I get have less than 5 tasks, are brand new rigs, or have RACs of less than 10..because of one/both of the previous two reasons.

Which bolsters my suggestion that MB credits be raised.......to get some of us AP enthusiasts to allow their rigs to crunch down the MB overload......

That would askew the credit system further away from the standard. It would make more sense to lower the AP credit so that it was more in-line with the standard as well, since it currently pays more when it obviously shouln't.

---

I will agree with AP being an opt-in sort of deal, but after looking at one of the causes of all of my pending credits/tasks.. AP should not be allowed on hosts that have less than a 250 RAC. That requirement makes sure they're going to actually try to do some crunching. 90% of the wingmen I get have less than 5 tasks, are brand new rigs, or have RACs of less than 10..because of one/both of the previous two reasons.

Which bolsters my suggestion that MB credits be raised.......to get some of us AP enthusiasts to allow their rigs to crunch down the MB overload......

That would askew the credit system further away from the standard. It would make more sense to lower the AP credit so that it was more in-line with the standard as well, since it currently pays more when it obviously shouln't.

The old who pays what syndrome, eh?
We really don't wanna go down that sorry old road again....

I will stand back and crunch what comes down the pipe.......

---

"Time is simply the mechanism that keeps everything from happening all at once."