When I check my stats, i go up...then down...
users seem to be increasing..or is this just another
someone with lots of unused accounts?
Is 60,000 users a reasonable number?

> When I check my stats, i go up...then down...
> users seem to be increasing..or is this just another
> someone with lots of unused accounts?
> Is 60,000 users a reasonable number?
>
Yes, I guess those numbers are some what correct. The number of hosts I don't trust very much since I've seen accounts with a huge number of computers that can't be correct.
All the numbers comes from the project directly (Statistics for SETI@home is updated every 12 hours) and you can find the XML-files here: http://setiweb.ssl.berkeley.edu/sah_stats.php

Happy crunching!

---

Happy crunching

I bet at least 20% have not crunch one BONIC/SETI unit yet....

Hey Voyager is that an original T-Shirt cool....


Timmy

---

> Yes, I guess those numbers are some what correct. The number of hosts I don't
> trust very much since I've seen accounts with a huge number of computers that
> can't be correct.
actually some people run pretty huge crunching farms... rooms filled with bare motherboards.

My stats site reports the following:

Total number of teams in database: 10343
Total number of users in database: 60169

From http://stats.kwsn.net/teams.php?proj=sah

The XML I get from berkeley only contains people who have received credit from at least 1 work unit (possibly 0.001 credits). By sorting the user listing (http://stats.kwsn.net/users.php?proj=sah) by the "Last result received" column and doing some URL mangling, it looks like roughly 28,000 users are inactive. (haven't returned a work unit in about 14 days). So that leaves about 32,000 active users right now.

FYI :)

---

A member of The Knights Who Say NI!
For rankings, history graphs and more, check out:
My BOINC stats site

> My stats site reports the following:
>
> Total number of teams in database: 10343
> Total number of users in database: 60169
> like roughly 28,000 users are inactive.

So that leaves about 32,000 active users right now.


Nice work Toby, I was being conservative with my estimate. It will be interesting to see what happens when all of SETI Classic converts and
how long some of the top 1000 crunchers tumble! :-(

I hope the team at Berkeley is working hard to ensure it will happen provided they’ll have all the equipment to handle the load...

Timmy

---

> > My stats site reports the following:
> >
> > Total number of teams in database: 10343
> > Total number of users in database: 60169
> > like roughly 28,000 users are inactive.
>
> So that leaves about 32,000 active users right now.
>
>
> Nice work Toby, I was being conservative with my estimate. It will be
> interesting to see what happens when all of SETI Classic converts and
> how long some of the top 1000 crunchers tumble! :-(
>
> I hope the team at Berkeley is working hard to ensure it will happen provided
> they’ll have all the equipment to handle the load...
>
> Timmy
>

If there are roughly 32k active SETI/BOINC users out of roughly 60+k converts, what do you think, say a conservative 1 million, more converts will do to the BOINC service? We've been running very well for over a month now without any major problems. I've been getting new WUs on a regular basis, when needed. I do not remember seeing an "Out of work" message from the SETI/BOINC server for quite some time.

Check out these stats I just got from SETI "Classic":

...................Total.............Last 24 Hours
Users:...........5266993........1336
Results rcvd:..1660448763...1445771

With almost 1.5 million results returned in the last 24 hours, I believe it would be safe to say that at least 1 million users returned those results.

I'm also running (well stalled) LHC. I have recieved 1 WU from them in the past 3 weeks. Their servers have been down for the past 3 to 5 days. This project is new with nowhere near the number of active users as SETI/BOINC. Every time I go to the sight, when the site is up, they have a message that says the servers are up and low on work. Could we look forward to this type of service if 1+ million "Classic" users convert to BOINC? I am honestly dreading the day that "Classic" SETI gets shut down.

L8R....

---

---

CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath

> Check out these stats I just got from SETI "Classic":
>
> ...................Total.............Last 24 Hours
> Users:...........5266993........1336
> Results rcvd:..1660448763...1445771
>
> With almost 1.5 million results returned in the last 24 hours, I believe it
> would be safe to say that at least 1 million users returned those results.
>

Siran,

The stats you display show that 1,336 users returned those 1.5 million results. The number of active users (returned at least one unit in past month) on classic hovered between 500,000 and 600,000. Given the likelihood of ghosts in classic (machines that have been 'forgotten' but are still running the client), I would guess the number of new BOINC users when classic is ended to be more around 400,000 or so at most.

sorry , my mistake , double post , deleted



friendly and respectful
byron

> With almost 1.5 million results returned in the last 24 hours, I believe it
> would be safe to say that at least 1 million users returned those results.

I think you are over estimating a lot here... Any new-ish computer can do upwards of 8 work units per day. I'm sure classic still has a bunch of Pentium IIs crunching but I think there is enough new hardware running on it to say the actual number of users who returned a result in the last 24 hours is probably closer to 2-300,000. With some computers not running 24/7 and slower machines, that would probably be about right for the number of 'active' users to be at 450,000 wich is where it is right now according to http://setiathome.ssl.berkeley.edu/numusers.html.

> I'm also running (well stalled) LHC. I have recieved 1 WU from them in the
> past 3 weeks. Their servers have been down for the past 3 to 5 days. This
> project is new with nowhere near the number of active users as SETI/BOINC.
> Every time I go to the sight, when the site is up, they have a message that
> says the servers are up and low on work. Could we look forward to this type
> of service if 1+ million "Classic" users convert to BOINC? I am honestly
> dreading the day that "Classic" SETI gets shut down.

You must have missed the whole "LHC is going down for 2-3 months for upgrades". There might be occasional test work units but they are officially offline until January or so.

However your worries about BOINC being able to support all the classic users are valid. BOINC will not send out every work unit as often as needed to fill demand like classic did (often 10+ times). Therefore the number of users it can handle will be greatly reduced. I believe I have heard Dr. Anderson say that seti@home only has enough work for about 200,000 users. This is one of the reasons they developed BOINC - to share the CPU power that was going to waste under classic. Signing up with several projects will most likely be the only way to keep your computer busy 24/7.

Just remember to check my stats site so you can keep track of your total stats easily :)

---

A member of The Knights Who Say NI!
For rankings, history graphs and more, check out:
My BOINC stats site

Any new-ish computer can do upwards of 8 work units per day
=========

Actually it's more like 16-20 per day, all of my computers put out at least 16, some of them more. Some of the 4 or 8 CPU Computers probably put out more than that even I would think...

---

> > Check out these stats I just got from SETI "Classic":
> >
> > ...................Total.............Last 24 Hours
> > Users:...........5266993........1336
> > Results rcvd:..1660448763...1445771
> >
> > With almost 1.5 million results returned in the last 24 hours, I believe
> it
> > would be safe to say that at least 1 million users returned those
> results.
> >
>
> Siran,
>
> The stats you display show that 1,336 users returned those 1.5 million
> results. The number of active users (returned at least one unit in past
> month) on classic hovered between 500,000 and 600,000. Given the likelihood
> of ghosts in classic (machines that have been 'forgotten' but are still
> running the client), I would guess the number of new BOINC users when classic
> is ended to be more around 400,000 or so at most.
>

Scott, uh, I believe I have to beg to differ. I believe that the 1336 users are new users signed up in the past 24 hours, and not the number of users that returned results in the past 24 hours. Hang on, let me go check it out again....

Yes, I was right. I got this text from the stats page. The stats, by the way, have not changed since I was there yesterday:

# The number of users in the past 24 hours represents the number of NEW users, not the number of users who have connected within 24 hours.

Let's use a conservative number of WUs per PC like 8. If those 1,445,771 WUs were returned by individual PCs and not "farms" of PCs, that would mean that 180,721.4 active users returned WU results. It would be nice to see stats given on the number of "real" active users on "Classic", and BOINC for that matter.

I had forgotten that I was sending in between 5 and 8 WU results per day before converting to BOINC. So, even if there are only 180+k active users on "Classic", it would still be a major impact to BOINC for the to convert.

L8R....

---

---

CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath

> > With almost 1.5 million results returned in the last 24 hours, I believe
> it
> > would be safe to say that at least 1 million users returned those
> results.
>
> I think you are over estimating a lot here... Any new-ish computer can do
> upwards of 8 work units per day. I'm sure classic still has a bunch of
> Pentium IIs crunching but I think there is enough new hardware running on it
> to say the actual number of users who returned a result in the last 24 hours
> is probably closer to 2-300,000. With some computers not running 24/7 and
> slower machines, that would probably be about right for the number of 'active'
> users to be at 450,000 wich is where it is right now according to <a> href="http://setiathome.ssl.berkeley.edu/numusers.html">http://setiathome.ssl.berkeley.edu/numusers.html[/url].
>

** Yes, you are correct in my over-estimation. I spaced on the fact that I was sending between 5 and 8 WU results per day. **

> > I'm also running (well stalled) LHC. I have recieved 1 WU from them in
> the
> > past 3 weeks. Their servers have been down for the past 3 to 5 days.
> This
> > project is new with nowhere near the number of active users as
> SETI/BOINC.
> > Every time I go to the sight, when the site is up, they have a message
> that
> > says the servers are up and low on work. Could we look forward to this
> type
> > of service if 1+ million "Classic" users convert to BOINC? I am
> honestly
> > dreading the day that "Classic" SETI gets shut down.
>
> You must have missed the whole "LHC is going down for 2-3 months for
> upgrades". There might be occasional test work units but they are officially
> offline until January or so.
>

** Ah, so that's why they're lagging. I do not spend a lot of time on their site. Thanks for informing me about it. **

> However your worries about BOINC being able to support all the classic users
> are valid. BOINC will not send out every work unit as often as needed to fill
> demand like classic did (often 10+ times). Therefore the number of users it
> can handle will be greatly reduced. I believe I have heard Dr. Anderson say
> that seti@home only has enough work for about 200,000 users. This is one of
> the reasons they developed BOINC - to share the CPU power that was going to
> waste under classic. Signing up with several projects will most likely be the
> only way to keep your computer busy 24/7.
>

** See my reply to Scott Brown.... **

> Just remember to check my stats site so you can keep track of your total stats
> easily :)
>

** No offense intended, but I have a BOINC Stats banner letting me know what my stats are. See below.... **

L8R....

---

---

CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath

> Scott, uh, I believe I have to beg to differ. I believe that the 1336 users
> are new users signed up in the past 24 hours, and not the number of users that
> returned results in the past 24 hours. Hang on, let me go check it out
> again....
>
> Yes, I was right. I got this text from the stats page. The stats, by the
> way, have not changed since I was there yesterday:
>
Actually, it is users that have received their first credit in the last 24 hours. Users that have not received credit do not appear in the XML files so they cannot be counted. I have watched LHC and noticed that days after the cutoff of new users, the number of new users is still creeping up slowly.

---

BOINC WIKI

Guess I am a bit overworked...I knew that about the users number and still typed my response (Sorry Siran and thanks JM VII).

I was mainly disagreeing with the 1.5 million esimate. Looking at the total users graph on SETI classic, there would appear to be a direct mapping of the loss of about 60,000 users since June and the number of SETI BOINC users. Thus, it would appear that the likely maximum number of users on BOINC SETI following the termination of classic would be around 600,000. There are two limitations to this one-to-one assessment, however. The 600,000 number could be a considerable overestimate if a significant number of machines still returning results are simply doing so on auto-pilot (e.g., a SETI user once worked at a place and left a few machines with SETI as a service, etc.). On the other hand, the news of the complete conversion to BOINC could bring in many new users (who never did SETI classic) or could reawaken some the the 4.5+ million non-active classic users, which would make 600,000 a substantial underestimate. The estiamte also ignores the possibility that many classic users will refuse (or be unable) to convert (e.g., they have machines without internet connections, etc.).

Of course, a very reasonable estimate could be obtained by simply surveying a random sample (say about 1,000 or so) of the classic e-mail accounts.

All of you who have mentioned that seti can't provide enough work for 600K users are correct...and also correct on one of the major reasons BOINC was created.

(the others being the ability to update worker code to do different science and send different sized WUs, etc)

CPDN could keep quite a large number busy, it would be interesting to ask them just how many WUs their servers can pop out.

LHC and predictor are down ATM...predictor *cough* predicts it will be up when it gets the database converted to the new servers (1 week+ and counting).

LHC has to fight with a Windows centric environment (Cern), and thus they were saddled with a poor server equipment (Dual Pentium III) for their intial test of the BOINC waters. The BOINC server code is written for [name your variety]-nix (Linux, Unix, etc.) O/S.

Einstein at home looks promising. Their programmers are active on the developer boards, and they are deciding on multi-CPU (Opteron vs. Xeon) servers...so they should have good capacity. How many WUs they can produce we don't know yet.

Folding @ Home has been in conversation with BOINC about that possibility...some testing was done by their professor/project head.

The Lattice project is in BOINC Beta right now, and will be producing a variety of science applications to release under BOINC. Information seems to imply they might also release Globus applications (another grid system) to BOINC users who elect to do this.

---

> > Scott, uh, I believe I have to beg to differ. I believe that the 1336
> users
> > are new users signed up in the past 24 hours, and not the number of users
> that
> > returned results in the past 24 hours. Hang on, let me go check it out
> > again....
> >
> > Yes, I was right. I got this text from the stats page. The stats, by
> the
> > way, have not changed since I was there yesterday:
> >
> Actually, it is users that have received their first credit in the last 24
> hours. Users that have not received credit do not appear in the XML files so
> they cannot be counted.
>

You failed to quote the piece of text I copied and pasted from the "Classic" site. It clearly stated that it IS the number of new users and NOT the number of users who have connected in the past 24 hours.

> I have watched LHC and noticed that days after the
> cutoff of new users, the number of new users is still creeping up slowly.
>

That's LHC and BOINC not SETI@Home Classic....

L8R....

---

---

CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath

> > I have watched LHC and noticed that days after the
> > cutoff of new users, the number of new users is still creeping up
> slowly.
> >
>
> That's LHC and BOINC not SETI@Home Classic....
>
> L8R....
>
> ---
>
My appologies, I thought this thread was about BOINC behavour.

---

BOINC WIKI

> > > I have watched LHC and noticed that days after the
> > > cutoff of new users, the number of new users is still creeping up
> > slowly.
> > >
> >
> > That's LHC and BOINC not SETI@Home Classic....
> >
> > L8R....
> >
> > ---
> >
> My appologies, I thought this thread was about BOINC behavour.
>

No applogy required. It is about BOINC, and the impact that the un-converted "Classic" users will have if they convert to BOINC. I dread the day. I believe I will probably get 1 or 2 WUs in a day or 2 to process when we gain that many more BOINC users.

L8R....

---

---

CAPT Siran d'Vel'nahr - L L & P _\\//
Winders 11 OS? "What a piece of junk!" - L. Skywalker
"Logic is the cement of our civilization with which we ascend from chaos using reason as our guide." - T'Plana-hath

> No applogy required. It is about BOINC, and the impact that the un-converted
> "Classic" users will have if they convert to BOINC. I dread the day. I
> believe I will probably get 1 or 2 WUs in a day or 2 to process when we gain
> that many more BOINC users.
>
Join several projects. I have joined several different projects. The computers that are capable of running CPDN will never run out of work again.

---

BOINC WIKI