I have just now read this thread after getting back from a business trip. I had not realized these frustrations.

I have been pulled in 2 directions; one to limit the # of servers and the other to expand the # of servers. I have chosen the latter. I am happy that we have just about cleared out 2 of them because ports 2000 and 3000 were only intended to be temporary to begin with. For NPLB to be popular and successful, it is my opinion that there should be many different choices for people. This is something that David (Ironbits) has mentioned also.

My dream 1: For someone to come here and have a whole plethora of different k-ranges, n-ranges, individual k's, etc. to choose from so that the project is as interesting as possible to the most # of people.

My dream 2: For the servers to be extensively tested before being loaded with "production" data and that everything works correctly the first time.

#2 did not happen at all on port G8000, which caused several problems. I made the assumption that it was properly tested because the code had been "copied" from another server...poor assumption on my part.

I must apologize for the lack of testing on the port. You have my word that before any new port is added in the future that proper test cases will be set up and conducted before any real data is run through them.


Gary

[quote=kar_bon;163106]look at gary's page for his GB4000 and GB8000 server:
there're results for GB8000 from 2009-02-13 on:

and

results_20090214_0001_GB_nplb_4000.txt is from GB8000!!!!


i tried to progress the new results but yesterday it was the [B]horror[/B]:

1.
results_20090212_0000_IB_nplb_4000.txt contains this:
[code]
user=gd_barnes
[2009-02-11 01:23:03]
511*2^630637-1 is not prime. Res64: 27C0C814EE688F0D Time : 2465.0 sec.
user=gd_barnes
[2009-02-11 00:40:34]
511*2^630637-1 is not prime. Res64: 27C0C814EE688F0D Time : 8458.0 sec.
[/code]2x the same result!!! how did this happen?

2.
results_20090214_0001_GB_nplb_4000.txt contains this:
[code]
user=MyDogBuster
[2009-02-13 00:00:22]
315*^600282-1 is not prime. Res64: 71F02131A1C1E24C Time : 467.0 sec.
[/code]this is a result from the server GB8000!!! again: how did this happen?

3.
results_20090213_0001_GB_nplb_4000.txt
results_20090214_0001_GB_nplb_4000.txt
are totally unuseable!!!

with "*^" instead of "*2^" and results not time-sorted!

so please to all:

check everthing first before setup a new server and test this!

to Gary:

please, 4 servers (3 for the Drives #5, #6 and #7 and one for other things) will be enough.
now we are at [B]8[/B] servers and with such sh*t resultfiles it cost me more time to correct those data.
NPLB -> No Prime Left Behind implies to be correct but please in all purposes!

hope this will be considered!

PS: i've uploaded the 3 above files corrected here: [URL="http://www.rieselprime.de/dl/new_res200902.zip"]www.rieselprime.de/dl/new_res200902.zip[/URL][/quote]
Oh man--of all the days for things to go wrong, just right after I got sick and couldn't get on the forum for three days! :rant:

I seriously thought these results files were all taken care of prior to my absence. Well, I was aware that they were sorted a little wrong, but I somehow thought they were still compartmentalized correctly within the correct dates. I'm still not quite sure why I thought that when in reality it turns out they were completely and totally gibberish! :rolleyes:

Anyway, thanks a bundle, Karsten, for cleaning up this mess. I'll get the corrected files uploaded to the [url]http://nplb-gb1.no-ip.org/llrnet/results/[/url] web server sometime later today. (And yes, I *will* be sure to write a note to myself about it so I don't forget. :smile:)

As always, thank you for keeping an ever-vigilant eye on these results files--it often saves a lot of mess later when I or anyone else involved in processing results messes up and lets the messed-up files slip through the cracks. :smile:

Max :smile:

[quote=mdettweiler;163298]Anyway, thanks a bundle, Karsten, for cleaning up this mess. I'll get the corrected files uploaded to the [URL]http://nplb-gb1.no-ip.org/llrnet/results/[/URL] web server sometime later today. (And yes, I *will* be sure to write a note to myself about it so I don't forget. :smile:)[/quote]
Okay, I've uploaded the fixed results files to the server and have duly verified that the fixed files are in place as needs be. Hopefully this will be the end of this mess. :smile:

The 'Stats by Server' report has been given some recent attention

[url]http://stats.ironbits.net/statsnew/stats_by_server.php[/url]

I've just noticed that on the Progress Report page, all of the columns for all days have everything up to and including the 10 A.M. hour zeroed out. I presume that this is a glitch of some sort, since the "total" columns for all the affected rows still appear to show the correct amounts (that is, including the real numbers behind the data that shows up as zeroed out)?

Edit: I've attached a screenshot that shows this more clearly.

Yes.
That's right
I am working on the back end of it to speed it up.
You are supposed to be asleep.
Close your eyes and think of your happy place.
Nothing to see here

:P

ok.

You can open them again now.

It now takes just 8 seconds to refresh all of the reports on the NPLB stats pages, down from about 4 minutes.

Alterior motive revealed:

Added the number of digits to the primes list
[url]http://stats.ironbits.net/statsnew/prime_list.php[/url]

Using Gary's accurate formula: int(n*log(2)+log(k)+1)

It will be added to the email notifications as an aid to help you estimate if your prime should be added to the top-5000 database.

PS - Gary's formula calculates 199527 digits for that prime, which is a fair bit more than the estimate of 86,653 that you gave.

*** sorry Gary - gone off topic here, please move these posts to the NPLB database thread. Thanks. ***

That's neat!

[QUOTE=IronBits;164064]That's neat![/QUOTE]

but not correct!!!

example:

893*2^639440-1 got 192494 digits, not 443234!!!

2^1000000-1 got 301030 digits, so a value your can easily remember:

about 300000 digits for n=1 million!

[b]PS: spot the error:

you used ln not log! logarithm to base 10! not 2,71828182845904 (=e)![/b]

thanks.
confirmed int(n*log(2)+log(k)+1) is not correct

I also used log10() and got the wrong answer. hmm. R&D in progress.