Searching for some beta testers after the easter weekend

ltd · 2006-04-06, 15:43

Hi,

my plan to test a new server for the llrnet starting after the easter weekend. There would run some k/n pairs for k not in the PSP scope. the size starts with around n=300000. The purpose of the testt is to check the stability of my setup before we move the production server over to the new platform. I would need in total about four PCs running 24/7 to have enough load to see if everything works fine. Please be warned that there is a chance that the system does not run stable enough and so your machine could run out of work.

No we can not use PSP double check tasks as the residues for the low n tests would not be comparable.

I will use my home machine so i am open for another three PCs. If we find a prime while doing the test it would be big enough to enter the top 5000.

Plaese send a post here if you are willing to help. I will send the contact information as soon as i have everything set up.

Lars

P.S. this will be the new server where the whole DB (and stats) should find their home later in the year.
P.P.S It will have a static IP so no more problems with no-ip

Citrix · 2006-04-06, 15:51

I thought be were done with double check to 400K?

ltd · 2006-04-06, 16:07

As said before the tests are NOT from the PSP k values.

Tests from DC that could be run with the llrnet 3.5 client would be larger then 1.3M. At this level the tests run to long to give enough load on the server for my intendet tests.

So i did set up and presieved some other k to have something available for such tests.
The K are:272341,279767,285473,287393,289171,294181,
294241,305147,306893,308423,312121,312863

Guess why i choose them?

Lars

Citrix · 2006-04-06, 16:31

Quote:

Originally Posted by ltd

As said before the tests are NOT from the PSP k values.

Tests from DC that could be run with the llrnet 3.5 client would be larger then 1.3M. At this level the tests run to long to give enough load on the server for my intendet tests.

So i did set up and presieved some other k to have something available for such tests.
The K are:272341,279767,285473,287393,289171,294181,
294241,305147,306893,308423,312121,312863

Guess why i choose them?

Lars

Thanks for the clarification. I know why you choose these k's. I will let other people guess.

edit: Now I have 1000+ posts.

edit: It is unlikely that you will find a prime since most of these numbers have already been checked in that range.

ttp://www.glasgowg43.freeserve.co.uk/siersurv.htm

ltd · 2006-04-06, 16:50

Oops i did not know that link.

And i have to check if i have to remove some of my k due to a prime already found.

Edit: 294241,306893,312863 are prime i did only look into a wrong internal list. They would have never send out as i really marked them prime

Citrix · 2006-04-06, 17:10

That page has not been updated for a long time, so I think the k's have been tested much deeper.

There are some problems here you could look at

http://www.primepuzzles.net/problems/prob_049.htm

ltd · 2006-04-06, 18:05

I think i will choose some other k values for the "official" test.
I did not recognize that the k are reserved by other people.

As i need some tests with relative low n i think i will use k=85013,91549,94373,98749,263329,943373,992731 instead.

These k are not reserved at the moment. I will do some presieving so that the beta test is no waste of time and resources.

Lars

Citrix · 2006-04-06, 18:26

You could sieve to find the largest twin prime. ( I could PRP some of these candidates)

ltd · 2006-04-06, 19:18

Using some k*2^n+1 for the tests make it easier to make all the script tests with the original scripts without endangering the PSP life data.

Lars

Citrix · 2006-04-06, 20:31

You can test k*2^n+1, ones that are prime can further be tested on the -1 side. (If you don't want to do this, that is fine with me too)

Citrix

Dagger · 2006-04-07, 04:06

I'm game. Would be glad to help test it.

2006-04-06, 15:43	#1
ltd Apr 2003 772₁₀ Posts	Searching for some beta testers after the easter weekend Hi, my plan to test a new server for the llrnet starting after the easter weekend. There would run some k/n pairs for k not in the PSP scope. the size starts with around n=300000. The purpose of the testt is to check the stability of my setup before we move the production server over to the new platform. I would need in total about four PCs running 24/7 to have enough load to see if everything works fine. Please be warned that there is a chance that the system does not run stable enough and so your machine could run out of work. No we can not use PSP double check tasks as the residues for the low n tests would not be comparable. I will use my home machine so i am open for another three PCs. If we find a prime while doing the test it would be big enough to enter the top 5000. Plaese send a post here if you are willing to help. I will send the contact information as soon as i have everything set up. Lars P.S. this will be the new server where the whole DB (and stats) should find their home later in the year. P.P.S It will have a static IP so no more problems with no-ip

2006-04-06, 16:50	#5
ltd Apr 2003 2²·193 Posts	Oops i did not know that link. And i have to check if i have to remove some of my k due to a prime already found. Edit: 294241,306893,312863 are prime i did only look into a wrong internal list. They would have never send out as i really marked them prime Last fiddled with by ltd on 2006-04-06 at 17:03

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Weekend GT730 Project	masser	GPU Computing	17	2016-11-15 01:55
MemtestG80 and MemtestCL: Memory Testers for GPUs	Jeff Gilchrist	Hardware	0	2010-08-19 16:32
I'm Dreaming of a White Easter	xilman	Lounge	8	2008-03-24 19:39
Happy Easter!	mfgoode	Lounge	4	2007-04-11 03:51
Attention PRP Testers (with SSE2 non-P4 machines)	axn	Sierpinski/Riesel Base 5	5	2006-10-12 05:08

2006-04-06, 15:51	#2
Citrix Jun 2003 7×233 Posts	I thought be were done with double check to 400K?

2006-04-06, 16:07	#3
ltd Apr 2003 2²×193 Posts	As said before the tests are NOT from the PSP k values. Tests from DC that could be run with the llrnet 3.5 client would be larger then 1.3M. At this level the tests run to long to give enough load on the server for my intendet tests. So i did set up and presieved some other k to have something available for such tests. The K are:272341,279767,285473,287393,289171,294181, 294241,305147,306893,308423,312121,312863 Guess why i choose them? Lars

2006-04-06, 17:10	#6
Citrix Jun 2003 65F₁₆ Posts	That page has not been updated for a long time, so I think the k's have been tested much deeper. There are some problems here you could look at http://www.primepuzzles.net/problems/prob_049.htm

2006-04-06, 18:05	#7
ltd Apr 2003 2²·193 Posts	I think i will choose some other k values for the "official" test. I did not recognize that the k are reserved by other people. As i need some tests with relative low n i think i will use k=85013,91549,94373,98749,263329,943373,992731 instead. These k are not reserved at the moment. I will do some presieving so that the beta test is no waste of time and resources. Lars

2006-04-06, 18:26	#8
Citrix Jun 2003 7×233 Posts	You could sieve to find the largest twin prime. ( I could PRP some of these candidates)

2006-04-06, 19:18	#9
ltd Apr 2003 2²·193 Posts	Using some k*2^n+1 for the tests make it easier to make all the script tests with the original scripts without endangering the PSP life data. Lars

2006-04-06, 20:31	#10
Citrix Jun 2003 1631₁₀ Posts	You can test k*2^n+1, ones that are prime can further be tested on the -1 side. (If you don't want to do this, that is fine with me too) Citrix

2006-04-07, 04:06	#11
Dagger Dec 2005 5₈ Posts	I'm game. Would be glad to help test it.