Fast Breeding (guru management) - Page 54

debrouxl · 2016-06-03, 19:36

Carlos: I'll eventually get to updating that page, yeah. Today, I fed 14e a bit (need to do more), and entered most of the pastebin IDs you posted in the other topic.

Others: as I wrote above, the future unified, improved version of the number queue management code is in a private Git repository on Github, with a general todo list for non-functional and functional changes, and some SQL code + todo list for the DB upgrade process.
The main technologies involved are PHP 5.x and MySQL 5.5 (used by BOINC), on Ubuntu 14.04 LTS. IIRC, the web server is Apache 2.x, but the queue management code doesn't care about that, as long as it somehow sets PHP_AUTH_USER (for the not-yet-written RBAC).

* who could lend a hand working on the code ? The next task is pretty much making a PDO stub, BOINC PDO work-alike, so that the web pages can actually be rendered.
* who has the infrastructure to build a dedicated testing environment, publicly accessible 24/7, on which everybody would be granted sufficient privilege ? The one I built for initial testing is on a server I use for something else as well, so I'm afraid that this won't do. I could rent another Scaleway C1 server or equivalent for a while, but I think I already have enough rented servers...
TIA

XYYXF · 2016-06-03, 21:05

С188_140_113 received a full t60. Poly from Gimarel: http://www.mersenneforum.org/showpos...&postcount=631

wblipp · 2016-06-04, 02:55

Two quartics and a quintic are ready now

21725473066772863950504173513098364324584931693593^5-1
99095361695961099223925080051226097068789794489357^5-1

1913049992229156211591^11-1

debrouxl · 2016-06-04, 08:23

I'm silly. We don't need a centralized, shared development and testing environment for the NFS@Home development code: every interested developer can leverage e.g. Vagrant or Docker on his/her own computer

fivemack · 2016-06-04, 18:37

I have trial-sieved wblipp's numbers and queued to 14e.

I notice the 14e management page is running rather slowly now there are so many numbers sieving.

Also comparing the fourteen polynomials from gimarel's message and will queue the best to 15e. Probably this will take until tomorrow morning.

debrouxl · 2016-06-04, 19:18

Alright. I'll throw some numbers from post #542 (currently the oldest post with pending numbers, AFAICT) in the mix. We don't have to start them right now.
EDIT: in the meantime, I've performed a pass through the LA reservations reservations, progress and results thread, and updated the 14e and 15e queues.

Today, I spent several hours making an automated test setup of the queue management code through Vagrant. I reached a state where

Code:

vagrant up --provision (--provider virtualbox)
vagrant ssh
(in the SSH session)
cd /vagrant
./postinst.sh

* starts from a Debian Jessie amd64 base (more lightweight than Ubuntu 14.04 amd64, differences are minor anyway);
* automatically installs all dependencies;
* erases and fills in a copy of the existing databases;
* starts building the new unified database;
* sets up HTTP Basic authentication on the management part of the infrastructure, which is the first building block for the RBAC;
* makes the PHP code accessible on port 8080 on the host. Said PHP code should be able to run, provided a stub of the BOINC DB API.

frmky · 2016-06-06, 06:42

70841_53_minus1 is causing an error on all Linux machines. I've canceled those workunits.

fivemack · 2016-06-07, 09:16

Getting 241^125-1 from http://www.mersenneforum.org/showpos...&postcount=554 ready for submission

XYYXF · 2016-06-21, 00:37

The xyyxf pool for 14e is almost empty, right?

C206_134_84
C211_148_132
C212_146_41
C223_130_112
C224_138_70

debrouxl · 2016-06-21, 21:01

From post #554:

Code:

7282882732751770301917311394675342168569930282640579^5-1 (quartic)
37141302977105581064921322418772202338727337015016057^5-1 (quartic)
230605827695659261374183298936900535402657822464812511^5-1 (quartic)

are easier by GNFS than by SNFS, especially the second one.

However, the composite cofactor of

Code:

1398036042749871729843146160774339503205345274956506531^5-1 (quartic)

is C193, and 14e can cope with this one using 31-bit LPs, so I've started it.

RichD · 2016-06-22, 15:00

Quote:

Originally Posted by debrouxl

From post #554:

Code:

7282882732751770301917311394675342168569930282640579^5-1 (quartic)
37141302977105581064921322418772202338727337015016057^5-1 (quartic)
230605827695659261374183298936900535402657822464812511^5-1 (quartic)

are easier by GNFS than by SNFS, especially the second one.

Are you sure?
Looking at the first one it is a SNFS(208) or GNFS(155).
NFS@Home uses 29-bit for SNFS(208) and GNFS(155) mostly uses 30-bit.

2016-06-04, 18:37	#588
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 2³×11×73 Posts	I have trial-sieved wblipp's numbers and queued to 14e. I notice the 14e management page is running rather slowly now there are so many numbers sieving. Also comparing the fourteen polynomials from gimarel's message and will queue the best to 15e. Probably this will take until tomorrow morning. Last fiddled with by fivemack on 2016-06-04 at 18:55

2016-06-04, 19:18	#589
debrouxl Sep 2009 977₁₀ Posts	Alright. I'll throw some numbers from post #542 (currently the oldest post with pending numbers, AFAICT) in the mix. We don't have to start them right now. EDIT: in the meantime, I've performed a pass through the LA reservations reservations, progress and results thread, and updated the 14e and 15e queues. Today, I spent several hours making an automated test setup of the queue management code through Vagrant. I reached a state where Code: vagrant up --provision (--provider virtualbox) vagrant ssh (in the SSH session) cd /vagrant ./postinst.sh * starts from a Debian Jessie amd64 base (more lightweight than Ubuntu 14.04 amd64, differences are minor anyway); * automatically installs all dependencies; * erases and fills in a copy of the existing databases; * starts building the new unified database; * sets up HTTP Basic authentication on the management part of the infrastructure, which is the first building block for the RBAC; * makes the PHP code accessible on port 8080 on the host. Said PHP code should be able to run, provided a stub of the BOINC DB API. Last fiddled with by debrouxl on 2016-06-04 at 19:54

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
System management notes	kriesel	kriesel	7	2020-10-21 18:52
Improving the queue management.	debrouxl	NFS@Home	10	2018-05-06 21:05
Script-based Primenet assignment management	ewmayer	Software	3	2017-05-25 04:02
Do normal adults give themselves an allowance? (...to fast or not to fast - there is no question!)	jasong	jasong	35	2016-12-11 00:57
Power Management settings	PrimeCroat	Hardware	3	2004-02-17 19:11

2016-06-03, 19:36	#584
debrouxl Sep 2009 977 Posts	Carlos: I'll eventually get to updating that page, yeah. Today, I fed 14e a bit (need to do more), and entered most of the pastebin IDs you posted in the other topic. Others: as I wrote above, the future unified, improved version of the number queue management code is in a private Git repository on Github, with a general todo list for non-functional and functional changes, and some SQL code + todo list for the DB upgrade process. The main technologies involved are PHP 5.x and MySQL 5.5 (used by BOINC), on Ubuntu 14.04 LTS. IIRC, the web server is Apache 2.x, but the queue management code doesn't care about that, as long as it somehow sets PHP_AUTH_USER (for the not-yet-written RBAC). * who could lend a hand working on the code ? The next task is pretty much making a PDO stub, BOINC PDO work-alike, so that the web pages can actually be rendered. * who has the infrastructure to build a dedicated testing environment, publicly accessible 24/7, on which everybody would be granted sufficient privilege ? The one I built for initial testing is on a server I use for something else as well, so I'm afraid that this won't do. I could rent another Scaleway C1 server or equivalent for a while, but I think I already have enough rented servers... TIA

2016-06-03, 21:05	#585
XYYXF Jan 2005 Minsk, Belarus 110010000₂ Posts	С188_140_113 received a full t60. Poly from Gimarel: http://www.mersenneforum.org/showpos...&postcount=631

2016-06-04, 02:55	#586
wblipp "William" May 2003 New Haven 2·7·13² Posts	Two quartics and a quintic are ready now 21725473066772863950504173513098364324584931693593^5-1 99095361695961099223925080051226097068789794489357^5-1 1913049992229156211591^11-1

2016-06-04, 08:23	#587
debrouxl Sep 2009 977 Posts	I'm silly. We don't need a centralized, shared development and testing environment for the NFS@Home development code: every interested developer can leverage e.g. Vagrant or Docker on his/her own computer

2016-06-06, 06:42	#590
frmky Jul 2003 So Cal 2⁴×7×19 Posts	70841_53_minus1 is causing an error on all Linux machines. I've canceled those workunits.

2016-06-07, 09:16	#591
fivemack (loop (#_fork)) Feb 2006 Cambridge, England 2³×11×73 Posts	Getting 241^125-1 from http://www.mersenneforum.org/showpos...&postcount=554 ready for submission

2016-06-21, 00:37	#592
XYYXF Jan 2005 Minsk, Belarus 2⁴×5² Posts	The xyyxf pool for 14e is almost empty, right? C206_134_84 C211_148_132 C212_146_41 C223_130_112 C224_138_70