It's time to bight the bullet, shutdown SETI classic and get all that hardware being used for an out-of-date project across to SETI BOINC. Yeah there will be hiccups with the transition but it happens and we get over it. I can't see how Berkeley can justify keeping classic going anymore. Come on bring it on!

Live long and crunch.

---

Paul
(S@H1 8888)
And proud of it!

I absolutely agree. It is time to make the switch. And while your at it let’s shut down the Seti enhanced site and start up that software here. It’s time for both to happen. Even if the project has to shut down for a few days to accomplish it let’s do it! If you agree + rate the above post!

---

Boinc....Boinc....Boinc....Boinc....

Don't mean to sound rude, but it's bite not bight.

---

I'd give this thread 10 +'s if I could. Many of us have been saying this for months (at least weeks) but nobody is listening (enough to respond to our chants)

---

May this Farce be with You

Don't mean to sound rude, but it's bite not bight.

Paul was obviously thinking about the Great Australian Bite, where a long time ago, a mythical spirit from the dreamtime obviously got quite upset and took an awfully big bight out of South Australia and Western Australia :).

I absolutely agree. It is time to make the switch. And while your at it let’s shut down the Seti enhanced site and start up that software here. It’s time for both to happen. Even if the project has to shut down for a few days to accomplish it let’s do it! If you agree + rate the above post!

I don't quite think that _enhanced is quite ready for prime time. There are a few errors yet to be nailed down, in my opinion (namely the 'Max CPU time exceeded' error). However, as to ending Classic... Pull the plug already!!!

---

https://youtu.be/iY57ErBkFFE

#Texit

Don't blame me, I voted for Johnson(L) in 2016.

Truth is dangerous... especially when it challenges those in power.

I agree as well. I just took a look at the Server Status page, and I saw the most amazing thing. All were green!! Maybe something got the hamsters runnin! hehehe. Keep up the good work Berkley. And keep on crunching Crunchers!

Jeremy

---

I agree as well. I just took a look at the Server Status page, and I saw the most amazing thing. All were green!! Maybe something got the hamsters runnin! hehehe. Keep up the good work Berkley. And keep on crunching Crunchers!

Jeremy

Availavle work also went up, but "waiting for validation" when up by 3x the available work.

The bottle neck just shifted, It's not gone.

---

Eh... if Berkeley actually had solid time (meaning no snafu's or other errors) I am POSITIVE that the bottleneck will correct itself. The servers can only crunch so fast, and trying to keep up with the extremes of creating, distributing, and then verifying after an outage, personally, I'd say this last outage just proved that Berkeley is doing a GREAT job so far!

---

"Fiction reveals truth that reality obscures."
-Jessamyn West

Seems that is the case -- at best it seems that SETI BOINC is simply running very close to the edge. Essentially that means unless everything is running 'hot, straight and normal' and doing so for an extended period of time (measured at least in consecutive weeks, not days or hours), *something* will be 'bottlenecked'.

An important question is what can be done about that -- I suspect the Berkeley folks are cogitating on just that.

Available work also went up, but "waiting for validation" when up by 3x the available work.

The bottle neck just shifted, It's not gone.

---

I can't figure out how anybody could be POSITIVE the bottleneck will sort itself out in finite time. Look at the performance for the last six weeks or more. Computers shifting, networks reconfigured, processes re-distributed are the norm. Matt L must be pulling his hair out.

I think the best policy is to immediately begin to wean classic in order to stepwise move hardware over to boinc.

(By the way, I'd love to see command central describe their plan/strategy. At least we would 'know' what is likely to happen.)

---

May this Farce be with You

I can't figure out how anybody could be POSITIVE the bottleneck will sort itself out in finite time.

You've been around long enough that you should have seen this before.

Servers come back up, the BOINC clients report in (some of them with "help" from the users) the queues grow quickly as finished work comes flying in.

This time, we didn't see the usual impact on the upload/download servers and we didn't see big problems connecting to the scheduler.

The transitioner queue seems to be getting smaller.

Each time a queue empties, the resources used to process that queue become available for other tasks.

If I had one gripe, it is that the status page reports levels and not rates.

---

I agree, we need rates.

I disagree. The post I referred to implied that things will sort themselves out without human futzing around. Not true, not yet.

We need to shut down classic now and get the horsepower reassigned to boinc.

---

May this Farce be with You

We need to shut down classic now and get the horsepower reassigned to boinc.

To that end, I have an idea. Everyone one of us who thinks this transition should occur now rather than later, try the following, which I call Distributed E-Mailing (DEM).

Go to the setiClassic web and find a few addresses of users who are still returning wu's. Send a polite request to these folks to move over to boinc now for the good of both projects. This may tip the scales and accelerate the transition!

It isn't spam, and is intended to encourage those who are faint of heart to make the switch.

---

May this Farce be with You

I agree as well. I just took a look at the Server Status page, and I saw the most amazing thing. All were green!! Maybe something got the hamsters runnin! hehehe. Keep up the good work Berkley. And keep on crunching Crunchers!

Jeremy

Availavle work also went up, but "waiting for validation" when up by 3x the available work.

The bottle neck just shifted, It's not gone.

Yeah I saw that, its up to over 570K now waiting to validate, but everything is still green. Might have to get those hamsters into over drive. I dunno.

Jeremy

---

I disagree. The post I referred to implied that things will sort themselves out without human futzing around. Not true, not yet.

But, does the necessary "futzing" take place because the forum members demand it, or does it happen because Matt and Bobb and the rest of the folks running the show look at the status page, or maybe even have something that we can't see?

---

Take a look at the Classic Bulletin Boards and I think you will find a strong sentiment against moving to BOINC/SETI. In fact, I have seen posts there that are rallying support for continuing "Classic" forever. We who have come over should at the least acknowlege the diehard Classic users, as it seems the folks at Berkeley are doing.

They need to find some way to give processing priority, (including efficient file system access), to the backend processes, ahead of the front end, generation of new work.

This would automatically throttle the system, to the max sustainable throughput, for the given amount of processing power, to do the validation and assimilation.

Giving priority to generation of new WUs, as they seem to have it now, is guaranteed to choke on the slow post processing steps, and the units waiting to validate just keeps going up and up and up.

They need to throttle the output of new WUs, until the backend is cleared out. It clearly isn't working to keep doing this.

Don

---

We need to shut down classic now and get the horsepower reassigned to boinc.

To that end, I have an idea. Everyone one of us who thinks this transition should occur now rather than later, try the following, which I call Distributed E-Mailing (DEM).

Go to the setiClassic web and find a few addresses of users who are still returning wu's. Send a polite request to these folks to move over to boinc now for the good of both projects. This may tip the scales and accelerate the transition!

It isn't spam, and is intended to encourage those who are faint of heart to make the switch.

From Another Forum Message:

NEWS FROM THE SETI@HOME STATISTICS PAGES

Seti@Home Classic Crunchers are closing in on the major milestone of TWO BILLION work units crunched. As of 08/01/2005 these dedicated participants in the worlds largest distributed computing project have produced 1,966,174,512 work units of processed work.

Projections based on emperical evidence indicate that these crunchers will reach the [bold]TWO BILLION milestone in mid September, 2005[/bold]. The hard core crunchers are producing nearly 800,000 work units per day. Seti@Home statistics show that there are approximatily 290,000 active crunchers working on the classic project. Many have been with the project for over four years, and still many in recent years. Active sign up for the classic project was ended on June 1, 2005.

Stay tuned for more news as the project approaches this major milestone.

I have been crunching SETI@Home since June 1999, now over 6 years ago. Several months ago, I read about the transition to BOINC/Seti. I have read these forums daily since then and keep up on all the news, updates, optimized clients, outages and bottlenecks. I installed BOINC/Seti on all my machines and crunched a few numbers to ensure that they all worked, then stopped. They are all ready for the final shut down of Classic and the transition to BOINC/Seti.

While, I agree that Seti Classic should be shut down, it doesn't appear that it will be until mid-September at the earliest. There is too much at stake. It's not about you; it's not about me; it's not about moving resources from one project to another. It's about marketing and milestones. There will probably be a lot of publicity and media attention when that TWO BILLION milestone in mid September, 2005 is reached. That equates to dollars, pure and simple. Dollars to spend on BOINC/Seti, new hardware, new resources and new projects.

Please don't alienate yourself and others by sending e-mails, on your own, directly to other participants who choose to stay with Classic until the end before making their transitions. Or if you choose to continue that avenue, which is your right, at least ask one or two of the people who actually work for the project, for their opinion before doing it. I for one would not like to receive several e-mails from several other participants asking me to move over. I already have; I'm ready for the shut down of Classic. I'm trying to help it along.

If you really, really and truly, wish to 'tip the scales and accelerate the transition' as you put it, convince as many people as you can right now to stop crunching BOINC/Seti and reutrn to Seti Classic for a week or two to accelerate the arrival of the TWO BILLION milestone sooner than mid-September, then Seti Classic can be turned off and then BOINC/Seti will be the only project with milestones to acomplish.

It doesn't mean you have to remove BOINC/Seti from your computer. Just take a short break from it and help 'finish Classic off' that much sooner. Otherwise, all the whining and cheerleading in the world won't shut Classic down before mid-September.

This is just my opinion. It's neither positive or negative; I'm neither whining nor cheerleading. After reading these forums for at least the past 6-7 months, I know better than to do any of that. Just my observations. I will sit back and watch the flames that I am sure will occur. Be gentle, I do not post much, if at all, just for this reason. (But I assure you, I read each and every post, even if it doesn't directly pertain to me and my setup and operation in this thread and all the others.)

Respectfully Submitted,
mjmanix
Metairie, Louisiana, USA

---

OS:WinXPPro SP2
CPU:Intel P4HT 3.2GHz
MB:ASUS P4P800-E
RAM:Corsair 2048MB PC3200DDR
HDD0:WDC 40GB P-ATA(OS, AV, Firewall, BOINC 5.10.16, etc)
HDD1:Maxtor 200GB S-ATA(non-essential files/program installers)
VGA:ATI A-I-W 9800Pro AGP 8x 128MB-DDR

They need to find some way to give processing priority, (including efficient file system access), to the backend processes, ahead of the front end, generation of new work.

This would automatically throttle the system, to the max sustainable throughput, for the given amount of processing power, to do the validation and assimilation.

Giving priority to generation of new WUs, as they seem to have it now, is guaranteed to choke on the slow post processing steps, and the units waiting to validate just keeps going up and up and up.

They need to throttle the output of new WUs, until the backend is cleared out. It clearly isn't working to keep doing this.

Don

In the past, the very loudest complaints are when people can't get new work -- when their crunchers are idle.

---