Posts by ivan

log in
21) Message boards : Number crunching : Best performing hardware (Message 1602280)
Posted 18 Nov 2014 by Profile ivan
Correct me if I'm wrong, but AFAIK the important thing for SETI is single precision performance.

The Xeon Phi 7120P seems to have a theoretical single precision peak of 2.4 TFLOPS/s:

A GTX780 or a Titan seems to be much faster:

The GTX 780 still offers respectable single precision performance though, clocking in at 4 Teraflops compared to the Titan's 4.5 Teraflops.

The Titan Black is rated at 5.1 TFLOPS/s

The GTX980 should be about the same with ~5 TFLOPS/s

I have a Xeon Phi. I gave up on trying to port BOINC and S@H to it. For the best performance you have to run native code on the Phi cluster (otherwise communication bottlenecks between the host and the cluster slow you down). That would mean running BOINC on the Phi as well and BOINC code is not the most portable (but not too bad if you ignore boincmgr and use boinccmd for all interactions).
Then there's the memory problem -- our particular model only has 8 GB of ram, which has to contain the OS as well as applications. At the moment top reports 7.6 GB free, so for 60 cores x 4 threads that'd be only around 30 MB available per thread; currently on an Ubuntu box s@h reports 104 or 164 MB virtual memory per process, 40 or 96 MB resident per process, and 12 MB KB shared. (Similar figures on a RHEL box; the one job on Ubuntu that's taking more RAM is not the vlar WU running at the moment.)
This might all change with the new Phis, which slot into motherboard sockets and can access main memory, but Intel hasn't offered me one to play with yet...
22) Message boards : Number crunching : How to limit size of (Message 1601179)
Posted 14 Nov 2014 by Profile ivan
Mine is 41 MB and 385000 lines

Over here they call this "willie waving"... :-)

[eesridr:BOINC] > wc 469562 5165182 51731379 [eesridr:BOINC] > head 1273852033 ue 167272.344521 ct 7.204450 fe 167272344521056 nm 10fe07ah.31298.5798.12.10.173_0 et 9.966059 1273852035 ue 167272.344521 ct 6.804424 fe 167272344521056 nm 10fe07ah.31298.5798.12.10.191_0 et 8.589399 1273855954 ue 167272.344521 ct 3890.751156 fe 167272344521056 nm 10fe07ah.31298.5798.12.10.100_0 et 3934.674194 Convert Unix timestamp to Readable Date/time (based on seconds since standard epoch of 1/1/1970) UNIX TimeStamp: 1273852033 Fri, 14 May 2010 15:47:13 GMT
23) Message boards : Number crunching : Lunatics Windows Installer v0.43a Maintenance Release (Message 1601095)
Posted 14 Nov 2014 by Profile ivan
Is there a way to add all 3 cuda apps, and can this be done by editing the app_info script since you only have an option to check one of the boxes?

As far as I understand it, you don't want to do that. The stock app sends the different versions because it doesn't know what hardware you have; once enough units have been crunched it selects the most efficient and only targets that version (the work-units are the same, it's just the targeted version that can vary). By selecting the most appropriate version during Lunatics install you can short-circuit this lengthy and inefficient selection process.
24) Message boards : Number crunching : BOINC 7.4.27? (Message 1600164)
Posted 12 Nov 2014 by Profile ivan

Download BOINC for Linux x64 (3.25 MB) BOINC 7.2.42
25) Message boards : Number crunching : tripple gtx 980 water cooled setup (Message 1599822)
Posted 11 Nov 2014 by Profile ivan
Hopefully that will solve their chronic fan-quality problems! :-b
26) Message boards : Number crunching : hyperthreading and processor affinity in linux with taskset (Message 1596056)
Posted 2 Nov 2014 by Profile ivan
So I moved it to local storage ( SSD ). Then looked a little closer at it while it ran. Regularly ( every few seconds ) hitting 100-200 MB/sec of disk IO ( over 1/2 second averages as viewed via iostat/iotop). So this project was IO bound. Kept my network and server overtasked.

I've installed lm-sensors and gkrellm on as many of my Linux boxes as I can. Then run sensors-detect to collect sensor info and run gkrellm, configuring which sensors and subsystems to monitor. I have a script that starts up
ssh -Yf user@machine gkrellm
for all my machines and then I can monitor them all in one workspace of my (Windows+cygwin) desktop. It's getting difficult to monitor those 20-CPU Xeons with only a 1920x1200 monitor, tho'but! :-/
27) Message boards : Number crunching : Panic Mode On (91) Server Problems? (Message 1593587)
Posted 28 Oct 2014 by Profile ivan
It seems to take a long time for the database engine to restart - whether it's loading tables into memory, integrity-checking the indexes, or something else, I don't know. But after a few hours, it lets out an almighty sneeze, and suddenly everything works at full speed again.

Yeah, response is suddenly back to normal now, database queries are up, and reported delays are more like usual.
28) Message boards : Number crunching : Panic Mode On (91) Server Problems? (Message 1593580)
Posted 28 Oct 2014 by Profile ivan
If you're only reporting or requesting work you get through (though I've yet to get any new work), but if you do both at the same time you get the HTTP error response.

Looks like I've managed to get one MB and ten APV7 tasks tonight (all AP tasks on my work laptop).

Other than that the SSP updates the page time yet the other numbers don't seem to be changing at all.

Seems to be some problems with the database, e.g. I get no response when I try to look at my Errors page to see if there are any new errors.
29) Message boards : Number crunching : Panic Mode On (91) Server Problems? (Message 1592535)
Posted 26 Oct 2014 by Profile ivan
Cricket just took a huge dive. Expect snow, hail, and thunderstorms.

(Link added.)
30) Message boards : Number crunching : Problems with 10.25.2014 (Message 1591862)
Posted 25 Oct 2014 by Profile ivan
Couple of weeks ago my Uni account started blocking several threads here on account of some metric or other. Curiously, most were threads started by Merle...
31) Message boards : Number crunching : intel i7 4790K and intel's warranty (Message 1591298)
Posted 24 Oct 2014 by Profile ivan
How long did you go with the stock cooler and how long before it was hitting such high temps? Thanks

Heh, not much longer than it took me to get eBuyer to deliver an after-market cooler. :-) I don't recall exactly how long it took to start bouncing off the rev-limiter when I fired up S@H, but I believe it was measured in seconds rather than minutes...
32) Message boards : Number crunching : intel i7 4790K and intel's warranty (Message 1590903)
Posted 23 Oct 2014 by Profile ivan
I suppose the best thing for me to do would be to put in cooler master and then if it is too loud I could go back to the intel heatsink/fan.

Question: Until then, what do I watch out for re. damage to MB or other components other than just watch temps for cpu & gpu cores? If anything?

Coming late to the party... Last year I installed a 4770K with the stock cooler. Under Mint Linux running 8x S@H MB jobs, it would hit 100 C and then throttle back to 3.5 GHz. I installed an aftermarket cooler and it then ran at full 3.9 GHz and is currently running core temperatures of 75-80 C. It's been doing that for nearly a year now.
So just keep an eye on the temperatures (and clock speed if you have a programme to monitor that).
33) Message boards : Number crunching : Panic Mode On (91) Server Problems? (Message 1590111)
Posted 22 Oct 2014 by Profile ivan
Summats up:-)

22/10/2014 11:19:36 | SETI@home | Scheduler request completed: got 107 new tasks

Got a fair stack of work including 3 APv7's

Swings & Roundabouts I guess.

Yeah, one of my new 20-core machines ran out and the other was on its last WU when the first got 98 new tasks and the second got 38.
22-Oct-2014 10:55:30 [SETI@home] Scheduler request completed: got 98 new tasks 22-Oct-2014 11:33:59 [SETI@home] Scheduler request completed: got 38 new tasks
34) Message boards : Number crunching : how many ways to fry a motherboard? (Message 1588558)
Posted 18 Oct 2014 by Profile ivan
Good to know. Thanks

The word we use in the states is usually Q-tip but I think we have other names for it too.

"Cotton bud" most places I've been.
As for spreading heat-sink compound, I keep all my old credit/debit cards and similar and use them to carefully smear the compound all across the surface of the CPU/GPU housing. (They are also handy in breaking apart pressed-together casings on all-in-one PCs, tablets, tc.)
35) Message boards : Number crunching : New PSU blows my mind. (Message 1588231)
Posted 17 Oct 2014 by Profile ivan
Ok, so now I have connected the 24 pin connector to the MB.

Next I know I need to connect an 8 pin ATX12V1 to the MB.

I have 2 cables that have 8 pin connections on each end - they have the label "CPU"

I think I have to connect one of these to that ATX12V1 on the MB


and probably to the "6+2 PCIE & 4+4 CPU" group.

If you mean that as sockets on the power supply, yes. That "group" would be to supply extra 12 V to the ATX input and directly to PCIE GPUs, etc (some have 6-pin inputs, some have 8-pin).
[Edit] Ah, yes, found a picture on eBuyer. Definitely connect to that group. In fact, you can hardly go wrong -- if it's 8-pin it goes there, if it's 6-pin it goes in the peripheral group. Ignore the CLINK unless you have a second P/S... [/Edit]

I know of course that I first have to install the cpu and heatsink/fan but as far as the PSU set up is that all I need to start. That is the 24 pin and 8 pin connection? The 16GB of memory is already set up on the MB too.

I also need to confirm that I am setting up that 8 pin correctly.

That should be enough to get a bios display, assuming you have an on-board VGA/video output.
36) Message boards : Number crunching : Panic Mode On (90) Server Problems? (Message 1585052)
Posted 11 Oct 2014 by Profile ivan

Patience, Julie. I'm just now starting to see an uptick in my home machine and overall RAC. Perhaps the APv7 results are starting to have an effect. Or maybe it's just a statistical blip...
37) Message boards : Number crunching : AP V7 (Message 1584287)
Posted 10 Oct 2014 by Profile ivan
First one is complete, just awaiting my wingman.

I've had two complete and pending (3,900 secs each/400 secs CPU) and seven in progress on my "stock" Linux box.
38) Message boards : Number crunching : Mismatch in time elapsed values? (Message 1582727)
Posted 6 Oct 2014 by Profile ivan
I'm think to have that you must have Hypo Threading turned on . If you wish to speed things up Go into your BIOS and turn it off . That's not good running Seti . Or having it on with just a P4 .

OTOH, many users have reported a 25% or so productivity increase when using Intel Hyperthreading. That is, individual tasks take less than twice as long when running two per physical core.

It probably depends on various factors related to cache size, cache speed, main memory speed, etc. and the sensible way to determine whether to have it on or not is to actually see which way works best on the system. IOW, try each for long enough to see how many tasks get done per day.

Well, I've been running two identical machines 7371751 and 7371070 for more than a month now. Admittedly the one using hyperthreading has been crunching for one day longer than the other, but it has a 14-15% advantage over the non-HT machine in both total credit and RAC. Unfortunately the non-HT machine will have to be shut off tomorrow while an engineer investigates a small anomaly in its RAID controller, so it will slip behind a bit more. When we start using the HT machine in earnest, I'll have to turn off HT as we will then want the fastest individual processes, not the highest throughput.
39) Message boards : Number crunching : Panic Mode On (90) Server Problems? (Message 1581647)
Posted 3 Oct 2014 by Profile ivan
While my previous Front Page Mac is now in Full Free Fall;
AP Validation pending (491) · Valid (30)

All AstroPulse v6 tasks for computer 7042852
State: All (277) · In progress (90) · Validation pending (179) · Validation inconclusive (2) · Valid (6) · Invalid (0) · Error (0)
40) Message boards : Number crunching : Panic Mode On (90) Server Problems? (Message 1581398)
Posted 3 Oct 2014 by Profile ivan
They fixed the validator.


Sorry, Julie, but from where I sit it failed again after about 30 minutes. Waiting for validation rising again.

It looks like it hasn't completely gone OTL. Over the past hour my AP valid have gone from 130 to 147. That trickle could stop at any time I expect.

Yes, with more data this morning the slope of the graph is lower, so it may be still processing but at a lower rate than results are returned.

Previous 20 · Next 20

Copyright © 2015 University of California