Trial Factoring
log in

Advanced search

Message boards : Number crunching : Trial Factoring

Previous · 1 . . . 9 · 10 · 11 · 12 · 13 · 14 · 15 . . . 23 · Next
Author Message
Gigacruncher [TSBTs Pirate]
Send message
Joined: 28 Mar 20
Posts: 48
Credit: 8,419,360
RAC: 0
Message 6247 - Posted: 16 Apr 2020, 20:07:25 UTC - in response to Message 6246.

SRBase is now 10th in overall production for Trial Factoring :)

When will we be no. 1? :)

Follow this link and let's see how fast we can move to no. 1 in overall Trial Factoring: https://www.mersenne.org/report_top_500_custom/?team_flag=0&type=1001&rank_lo=1&rank_hi=15&start_date=1995-01-01&end_date=


Overall: https://www.mersenne.org/report_top_500_custom/?team_flag=0&type=0&rank_lo=1&rank_hi=25&start_date=1995-01-01&end_date=

k3ack3r
Volunteer developer
Volunteer tester
Send message
Joined: 7 Aug 19
Posts: 19
Credit: 10,641,118
RAC: 1
Message 6251 - Posted: 17 Apr 2020, 0:47:46 UTC

mfakto new version, maybe people can sanity check it.
http://www.filedropper.com/mfakto015pre7win64

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7140
Credit: 42,729,227
RAC: 734
Message 6252 - Posted: 17 Apr 2020, 15:24:30 UTC
Last modified: 17 Apr 2020, 15:38:20 UTC

new windows opencl app is up

Changes:
- checkpointing every 2min now (depends on your GPU)
- new faster app with RDNA

Edit:
some WUs have reached the returned errors of 10 in total, I will unassign this work later. After a short test the WUs are ok, maybe a bad luck of bad hosts.

mmonnin
Send message
Joined: 1 Feb 17
Posts: 22
Credit: 287,110,473
RAC: 2
Message 6261 - Posted: 18 Apr 2020, 13:21:02 UTC

Much faster on my R7. 5:40 to 4hr

Profile [B@P] Daniel
Send message
Joined: 8 Oct 16
Posts: 1
Credit: 17,684,783
RAC: 0
Message 6262 - Posted: 18 Apr 2020, 15:18:41 UTC
Last modified: 18 Apr 2020, 15:20:13 UTC

I also had problem with missing libcudart.so.10.1 library on CentOS 7. To install it in "official" way, first install cuda repo in your system, like described on https://linuxconfig.org/how-to-install-nvidia-cuda-toolkit-on-centos-7-linux. I used following command to do this:

yum install https://developer.download.nvidia.com/compute/cuda/repos/rhel7/x86_64/cuda-repo-rhel7-10.2.89-1.x86_64.rpm


Then you have to install cuda-cudart package - this is the one with required library:

yum install cuda-cudart-10-1-10*


This package registers its custom lib dir in /etc/ld.so.conf.d/, so dynamic linker is able to find library without further hacks. After I installed this package, GPU WUs started working properly.
____________

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7140
Credit: 42,729,227
RAC: 734
Message 6264 - Posted: 19 Apr 2020, 9:54:05 UTC

There is a higher download error rate with input files, you can ignore it.

mmonnin
Send message
Joined: 1 Feb 17
Posts: 22
Credit: 287,110,473
RAC: 2
Message 6265 - Posted: 19 Apr 2020, 11:22:27 UTC

It's hard to ignore when we can only have 1 task per host and its minutes between tasks.

KEP
Volunteer tester
Send message
Joined: 28 Nov 14
Posts: 88
Credit: 1,102,770
RAC: 2
Message 6266 - Posted: 19 Apr 2020, 11:31:44 UTC - in response to Message 6265.

It's hard to ignore when we can only have 1 task per host and its minutes between tasks.


Yes, it is.

Unfortunantly when Rebirther was doing some testing locally, in regards to remove the last traces of the projectname we were not allowed to use, something happened and that affected the prepared work, and does for now unfortunantly give us a higher download error rate, because the work that was prepared was deleted and the system has to recreate the workunits once more. If we increase the max from 1 to 2 or 3, it might result in weeks without new work for the highend GPU users.

We are sorry for the inconvienience and hope it all soon corrects itself.

On a possitive note, this will not happen again, once we start getting work from Mersenne.org, wich we hopefully soon will :)

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7140
Credit: 42,729,227
RAC: 734
Message 6267 - Posted: 19 Apr 2020, 14:44:11 UTC

I dont know if I can remove all the GPU72 stuff but didnt recognize on the testserver that I had access to the real database and removed all the unprocessed work of GPU72 thats why the download errors growed up. The new assimilator was already compiled and waiting now for the rest of the work.

In the next step I will change the subproject name and then running some test units. If all is going well we can start with real work.

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7140
Credit: 42,729,227
RAC: 734
Message 6268 - Posted: 19 Apr 2020, 14:45:37 UTC - in response to Message 6265.

It's hard to ignore when we can only have 1 task per host and its minutes between tasks.


This was only setup for the long WUs. For the tiny ones we will have 50 max per host.

Gigacruncher [TSBTs Pirate]
Send message
Joined: 28 Mar 20
Posts: 48
Credit: 8,419,360
RAC: 0
Message 6269 - Posted: 19 Apr 2020, 14:49:25 UTC - in response to Message 6268.
Last modified: 19 Apr 2020, 14:51:03 UTC

It's hard to ignore when we can only have 1 task per host and its minutes between tasks.


This was only setup for the long WUs. For the tiny ones we will have 50 max per host.


Which tiny ones are you talking about, 72 to 73 bits? If so looking forward to it.

zombie67 [MM]
Avatar
Send message
Joined: 4 Dec 14
Posts: 30
Credit: 1,005,778,104
RAC: 23,454
Message 6270 - Posted: 19 Apr 2020, 14:51:07 UTC

Unsent is now zero.
____________
Reno, NV
Team: SETI.USA

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7140
Credit: 42,729,227
RAC: 734
Message 6271 - Posted: 19 Apr 2020, 14:54:28 UTC - in response to Message 6269.

It's hard to ignore when we can only have 1 task per host and its minutes between tasks.


This was only setup for the long WUs. For the tiny ones we will have 50 max per host.


Which tiny ones are you talking about, 72 to 73 bits? If so looking forward to it.


we starting from smaller ranges. This could struggle the server, need to think about it. Maybe some minutes tasks to have time to setup very small WUs.

KEP
Volunteer tester
Send message
Joined: 28 Nov 14
Posts: 88
Credit: 1,102,770
RAC: 2
Message 6272 - Posted: 19 Apr 2020, 15:08:12 UTC - in response to Message 6269.

Which tiny ones are you talking about, 72 to 73 bits? If so looking forward to it.


We will be starting with 70 to 71 bit. Due to the very short runtime, it may be best for the server, if they are actually 70-72 bit. It is as simple as search and replace, before Rebirther creates new work. That small task of search and replace, will in itself increase the chance of a factor found from ~1.436% to ~2.872% and it will increase the testime by a factor of 3. Let's see what Rebirther decides to do, but it is very simple to run from 70 bit to 72 without messing anything up.

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7140
Credit: 42,729,227
RAC: 734
Message 6273 - Posted: 19 Apr 2020, 15:14:54 UTC - in response to Message 6272.

Which tiny ones are you talking about, 72 to 73 bits? If so looking forward to it.


We will be starting with 70 to 71 bit. Due to the very short runtime, it may be best for the server, if they are actually 70-72 bit. It is as simple as search and replace, before Rebirther creates new work. That small task of search and replace, will in itself increase the chance of a factor found from ~1.436% to ~2.872% and it will increase the testime by a factor of 3. Let's see what Rebirther decides to do, but it is very simple to run from 70 bit to 72 without messing anything up.


I prefer 68-69.

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 7140
Credit: 42,729,227
RAC: 734
Message 6275 - Posted: 19 Apr 2020, 15:29:25 UTC

Pls stop/abort all the current work for GPU72.

Profile PDW
Send message
Joined: 15 Oct 15
Posts: 41
Credit: 1,077,002,394
RAC: 38,199
Message 6276 - Posted: 19 Apr 2020, 15:31:33 UTC - in response to Message 6275.

Pls stop/abort all the current work for GPU72.

Can't, someone already killed them !

Gigacruncher [TSBTs Pirate]
Send message
Joined: 28 Mar 20
Posts: 48
Credit: 8,419,360
RAC: 0
Message 6277 - Posted: 19 Apr 2020, 15:35:26 UTC
Last modified: 19 Apr 2020, 15:41:45 UTC

Reb, you need to look at the page you linked, this one:

https://www.mersenne.ca/status/tf/0/0/1/10000

Try the 597,726 first from 72-73, if you jump to 1,386,782 (68-69 bits) it means this project results will only be helpful in 20 years time. My advice is simple: start on line 100M with the lowest bits you think you are comfortable, stop and go to line 200M, star from lower bits, and so forth. This way you will be helping gimps on short term.

Gigacruncher [TSBTs Pirate]
Send message
Joined: 28 Mar 20
Posts: 48
Credit: 8,419,360
RAC: 0
Message 6278 - Posted: 19 Apr 2020, 15:39:53 UTC - in response to Message 6275.

Pls stop/abort all the current work for GPU72.


You had clients preparing themselves for a boinc challenge, you said you were going to only made server changes when the server dries out. You should accept previous wus when returned by clients.

Profile PDW
Send message
Joined: 15 Oct 15
Posts: 41
Credit: 1,077,002,394
RAC: 38,199
Message 6279 - Posted: 19 Apr 2020, 15:44:05 UTC - in response to Message 6278.

Pls stop/abort all the current work for GPU72.


You had clients preparing themselves for a boinc challenge, you said you were going to only made server changes when the server dries out. You should accept previous wus when returned by clients.

Too late, having realised he had removed them from the server I went and found them still running on the client so aborted there also.

Previous · 1 . . . 9 · 10 · 11 · 12 · 13 · 14 · 15 . . . 23 · Next
Post to thread

Message boards : Number crunching : Trial Factoring


Main page · Your account · Message boards


Copyright © 2014-2024 BOINC Confederation / rebirther