Posts by Area 51


log in
21) Message boards : Number crunching : Back to the Future? (Message 1258387)
Posted 649 days ago by Profile Area 51
If you didn't have a fixed point of reference for time, the Message Boards would make no sense at all, well, even less than they do now! Now consider, when WUs are sent out and results received back....the servers don't know which time zone that each WU was sent to or what time zone the result came from. Don't forgot, that each WU has a completion time and date. Can you imagine the chaos that would ensue in all the databases, if the servers had to make corrections for time zones? This is 1st day kindergarten stuff, on Gallifrey - they like order in matters of time, there.




I can imagine no chaos from using the time zone of the location where the project is located. There is no difference between arbitrarily chosen "base" time zones.



OK, so what happens during switchover from Summer to Winter time? At the chosen time of transition, the servers go back in time 1 hour. Work could be returned 'before' it was sent.....

And the converse? Going from Winter to Summer? How about tasks timing out if they were being processed close up to their original deadline....

And lets not even start to consider timestamps on log entries etc.

Berkeley time varies - it is not constant. The solution? Use a time datum that avoids these and other issues by not varying with the seasons. There happens to be such a datum, its called UTC. That it doesn't coincide with 'local server time' is irrelevant to the project but its use precludes a whole bunch of issues that would cause serious headaches for any sysadmin.
22) Message boards : Number crunching : Kittyman......rig down..... (Message 1254311)
Posted 657 days ago by Profile Area 51


And then I have to get off of my butt some day fairly soon and figure out that bit about making a recoverable copy of the OS with all updates applied, because once support for XP is withdrawn, if something like this happens and I can no longer use Windows Update to get the Service Pack and other updates, she's all done.



Seriously take a look at the Acronis range of software. I bought a license for Acronis Migrate Easy to move from HDD to SSD - and I now use this for system backup. Included in Migrate Easy is a facility to clone drives - and it works well - really well.

I use the cloning facility to clone my Windows boot SSD to HDD periodically (I have BOINC installed on a separate SSD to my Windows boot SSD). If anything changes on my Windows SSD, I just clone to my backup HDD. With this setup, I can be back and running from a boot drive failure in less than a minute (assuming I'm around to tweak the boot order in the BIOS and re-connect the HDD cables) - with exactly the same settings as the last clone. I can then re-image the failed SSD from the backup HDD and its like nothing ever happened. Money well spent methinks...


23) Message boards : Number crunching : Anonumous host throwing only errors, 3223 right now (Message 1246566)
Posted 673 days ago by Profile Area 51
oh boy it looks like im a culprit now! i changed the app_info in my 3 computers that are running GT 430's so that they will run 2 tasks at a time. one of these computers is now spitting out errors from what i can see (computer 6606314). i live in an apt during the week and these computers are at my house so i don't have physical access to them at the moment. can somebody look into the errors that system is putting out and tell me if they know what seems to be the cause? i'd like to fix it ASAP, thanks.



You've immediately differentiated yourself form the 'others' by spotting your problem and asking for help ;-). Puts you orders of magnitude in front of those who shall not be menioned as far as I am concerned!

Posting your app_info.xml here would be a good start. Also, did you recently update your drivers and if so, exctly how did you go about it?
24) Message boards : Number crunching : Anonumous host throwing only errors, 3223 right now (Message 1245541)
Posted 675 days ago by Profile Area 51
Sad though this is (to the extent it makes a complete mockery of our efforts to produce good results), and given that the project has the capacity to black-list hosts should they want to, I can only assume that the impact of these hosts is considered largely insignificant. Either that, or the project just don't want to get into the business of micro managing individual hosts.....

The trouble with the quota system as it is, is that it doesn't really protect against invalids - as they don't materialise until quorum cannot be reached (by which time, a rogue host has downloaded another 1,000 tasks). Fixing that would alleviate a lot of the issues....
25) Message boards : Number crunching : Anonumous host throwing only errors, 3223 right now (Message 1242422)
Posted 682 days ago by Profile Area 51
From previous similar discussions - wasn't there some sort of a 'blacklist' capability within BOINC? Can't remember whether it operated at the host or account level, but I agree - something ought to be done about hosts/users who just spit out rubbish and squander resources!
26) Message boards : Number crunching : CLOSED* SETI/BOINC Milestones [ v2.0 ] - XXIV *CLOSED (Message 1240563)
Posted 685 days ago by Profile Area 51
Crashed through 60,000,000 BOINC sometime yesterday - rapidly approaching 30,000,000 S@H.
27) Message boards : Number crunching : What the heck has happen to workunit credits? (Message 1239616)
Posted 687 days ago by Profile Area 51
I used to get around 90 for each WU now with my WU with either Fermi or Cuda I only get 30 so I have lost two thirds of my credits I have only noticed this since the outage


I think we are definitely in the realm of unintended consequences.....
28) Message boards : Number crunching : What the heck has happen to workunit credits? (Message 1239519)
Posted 687 days ago by Profile Area 51
I think the servers messed up.

Look at reported times and credits.

http://setiathome.berkeley.edu/workunit.php?wuid=998136901



Good spot. At least is shows that x41x isn't the only application demonstrating these characteristics. Question is, how widespread is this issue......
29) Message boards : Number crunching : SAH On Linux (Message 1238548)
Posted 691 days ago by Profile Area 51
You know I gotta think that somewhere out there someone hasn't built the ultimate SETI@HOME OS which is essentially no OS. Start by building a Linux kernel with practically nothing in it. Eliminate the scheduler and make it real time. All of this stuff is pretty much off the shelf. Now just move all the Lunatics apps down into the kernel. Not for the general public, but for the hard core I think it would be a fun project. Probably end up maintaining it for the rest of your life though.

Doug



There is a RAM based linux distro here:

http://www.dotsch.de/boinc/Dotsch_UX.html

specifically for BOINC which can be configured to run from secondary storage.

It's a good start, but I'm talking all the way. Something so hard core that nothing will interrupt it from its I/O and computing activities. Something so perverse that its very existence is for SETI. I've got some spare disks. I might just do this at least up to the point of moving the Lunatics source into the kernel.



....so, even more bare bones that. Presumably more bones than anything then!!!!

I'm not sure how much you'd gain by binning the entire GUI and literally only compiling in the bits absolutely necessary - unless it was being run on a CPU with shed loads of cores so that all those litle gains really multiplied out, but yes, it would be an interesting little aside!
30) Message boards : Number crunching : SAH On Linux (Message 1238531)
Posted 691 days ago by Profile Area 51
You know I gotta think that somewhere out there someone hasn't built the ultimate SETI@HOME OS which is essentially no OS. Start by building a Linux kernel with practically nothing in it. Eliminate the scheduler and make it real time. All of this stuff is pretty much off the shelf. Now just move all the Lunatics apps down into the kernel. Not for the general public, but for the hard core I think it would be a fun project. Probably end up maintaining it for the rest of your life though.

Doug



There is a RAM based linux distro here:

http://www.dotsch.de/boinc/Dotsch_UX.html

specifically for BOINC which can be configured to run from secondary storage.
31) Message boards : Number crunching : Panic Mode On (75) Server problems? (Message 1237452)
Posted 693 days ago by Profile Area 51
...and uploads & reporting now functioning......
32) Message boards : Number crunching : Changes in BOINCstats Page (Message 1234631)
Posted 698 days ago by Profile Area 51
I agree with the horizontal scrolling and speed issues - still undecided whether the eye candy really annoys me or if I just haven't got used to it yet.
33) Message boards : Number crunching : Running multiple GPU tasks on one video card. (Message 1234208)
Posted 699 days ago by Profile Area 51
I do seti@home and Einstein@home.

Einstein@home has this neat thing in the project preferences where you can tell it how many instances of cuda task you want your video card to do.

seti@home does not have this feature at this time. Is there a howto somewhere that tells how to setup so that my video card does more than one cuda work unit?

I have looked at Lunatics download site. They got stuff, but things are a bit convoluted in there. I am looking at something cleaner.



you will need to edit your app_info.xml. Replace all instances of <count>1</count> with the reciprocal of how many tasks you want to run per GPU (ie, 0.5 for two tasks, 0.3 for 3 etc etc). Don't forget to stop your host processing whilst you do this.

I don't have app_info.xml. I am running unaltered boinc 7.0.25. That is why I am looking for a good howto.



Since your computers are hidden, and that was not mentioned.....

The more information you give, the more likely you are to get an appropriate response.
34) Message boards : Number crunching : Running multiple GPU tasks on one video card. (Message 1234194)
Posted 699 days ago by Profile Area 51
I do seti@home and Einstein@home.

Einstein@home has this neat thing in the project preferences where you can tell it how many instances of cuda task you want your video card to do.

seti@home does not have this feature at this time. Is there a howto somewhere that tells how to setup so that my video card does more than one cuda work unit?

I have looked at Lunatics download site. They got stuff, but things are a bit convoluted in there. I am looking at something cleaner.



you will need to edit your app_info.xml. Replace all instances of <count>1</count> with the reciprocal of how many tasks you want to run per GPU (ie, 0.5 for two tasks, 0.3 for 3 etc etc). Don't forget to stop your host processing whilst you do this.
35) Message boards : Number crunching : Welcome to the 13 year club (Message 1234158)
Posted 699 days ago by Profile Area 51
Today is my 13th anniversary in participating in SETI@HOME. It all started by reading an article online.



well done - 13 posts in 13 years!!!! Seriously though, congrats on your commitment - there aren't that many still around with that staying power.
36) Message boards : Number crunching : Almost time for the next outage isn't it ??? (Message 1232580)
Posted 704 days ago by Profile Area 51
Aaaannd we're back.


...nice and quick.
37) Message boards : Number crunching : Type of work units v credit (Message 1232457)
Posted 704 days ago by Profile Area 51
On my ATI PC I'm getting long and short WU's and also AstroPulse, the run times and credit for each type of WU is similar, EG: longs take 1000 to 1300 seconds for 100 - 140 credit.

But what's with these .vlar work units?

They take twice as long for the same or less credit?

Can I change a setting so I don't get these?

Thanks



.err no, part of the rich tapestry I'm afraid.
38) Message boards : Number crunching : Task takes too lomg error (Message 1232112)
Posted 705 days ago by Profile Area 51
I ran mixed speed GPUs on http://setiathome.berkeley.edu/show_host_detail.php?hostid=6379672 without getting this issue for 3 months but over the past few weeks it is happening far more. I have just written a small C program that updates all the <rsc_fpops_bound> values in the client_state.xml to be 25 rather than 10 times the <rsc_fpops_est> values which should resolve this, I just need to stop BOINC and run it every few days. It also resets the DCF to 1.0.

I think you can do similar with Fred’s rescheduler, but have never used it.

The real problem is that BOINC thinks all the GPUs in a system must be the same speed and unless this is fixed running mixed speed GPUs will have issues. The main one is that it's impossible to get a stable DCF. I just hope set@home starts using the <dont_use_dcf/> facility.


This sort of thing is a pet hate of mine. BOINC does not work at the correct granularity. Each processing resource should have its own set of values. I think the current situation is as a result of the knee jerk reaction when CUDA processing first came about (ie, make it happen). In the meantime, nobody has gone back and re-coded to do resource management properly. Frustrating, but I don't think it will change very soon - if at all.

I guess the real problem is the legacy apps some people may be running, such that if the structure of the XML files was changed significantly - the old apps still in use would break. This all seems to stem from a fundamental design decision years ago (pre GPU), in the sense that BOINC can't handle the current possibility that newer machines may have multiple processing resources, each with their own characteristics.

IMHO, BOINC needs a complete re-write - but that would be a huge task - and that's even less likely to happen!
39) Message boards : Number crunching : Actual core speed discrepancy...or just my stupidity? (Message 1228802)
Posted 712 days ago by Profile Area 51
I have a random question, but rather than start a new thread I'll just post it here:

I noticed that some of the top 20 computers in the statistics page have more processors than sounds right.

For example: GenuineIntel
Intel(R) Core(TM) i7-3930K CPU @ 3.20GHz [Family 6 Model 45 Stepping 6]
(12 processors)

and

GenuineIntel
Intel(R) Core(TM) i7 CPU X 980 @ 3.33GHz [Family 6 Model 44 Stepping 2]
(12 processors)

Arent those CPU's 6 core and 4 core? Each of the computer also has 6 GeForce GTX 590's, but I didn't think those were listed as processors. Like my C2D systems that have 1 GPU don't say "3 processors", they only say "2 processors" as is normal for C2D.

Anybody know why those systems would say 12 processors?




...hyperthreading for the CPUs, and dual GPU cards for the GPUS.....
40) Message boards : Number crunching : Hello Computer 6137511, control your machine (Message 1228756)
Posted 712 days ago by Profile Area 51
I'd join in with a pitchfork but before Jason developed fixes specifically for my dual 560ti rig, I was producing invalids/errors like it was my job.


...the difference being of course that you did something about it. What I can never fathom is why these people 'participate' in endeavors such as this. They obviously like the idea of 'participating' - even apparently to the extent of installing optimised apps, but then they do not appear to have a modicum of interest with regards to the rubbish they are producing. Is it a language thing? or just an ego thing - perhaps the latter.

Sad really, for the project as a whole and for those of us who try our absolute upmost to produce quality output at every level. I feel bad about 1 non -12 error....., but 11,000 invalids? - I'd be swinging in the breeze!!!!


Previous 20 · Next 20

Copyright © 2014 University of California