90 Hours to crunch 1 task? Help!

Message boards : Number crunching : 90 Hours to crunch 1 task? Help!
Message board moderation

To post messages, you must log in.

1 · 2 · Next

AuthorMessage
Profile Blanchjoe

Send message
Joined: 27 Nov 02
Posts: 12
Credit: 7,148,533
RAC: 5
United States
Message 321327 - Posted: 31 May 2006, 14:09:17 UTC

Hello Everyone, I just recently moved to Bionic 515 and I just noticed that my last three WU took 70 and 90 hours to run? Am I reading the Bionic Manager correctly? Have the WU gotten significantly larger? Can someone let me know if this is normal. My system ran slow in the past (I was using an optimized Bionic 411) but it took no more than 16 hours to run a WU. I have Number of CPU's is 1 at 275 float point MIPS on Win98 with a 333mhz processor. Thanks! Blanchjoe
ID: 321327 · Report as offensive
Profile Pooh Bear 27
Volunteer tester
Avatar

Send message
Joined: 14 Jul 03
Posts: 3224
Credit: 4,603,826
RAC: 0
United States
Message 321330 - Posted: 31 May 2006, 14:15:31 UTC

The units are the same size, the application does a deeper scan, so yes with your P2, you will see this. There are varied hours, so some will be faster, some slower.



My movie https://vimeo.com/manage/videos/502242
ID: 321330 · Report as offensive
Profile Blanchjoe

Send message
Joined: 27 Nov 02
Posts: 12
Credit: 7,148,533
RAC: 5
United States
Message 322976 - Posted: 2 Jun 2006, 3:59:37 UTC

Thanks Pooh, Egoicly I knew I could never keep up with the mighty crunchers, but at least I felt that I could push a WU out of my poor 333 P2 in 18 hours, but now to see that it takes 70 or 90 hours is deflating needless to say. Ah me, well thanks for your insight. Blanchjoe
ID: 322976 · Report as offensive
Alinator
Volunteer tester

Send message
Joined: 19 Apr 05
Posts: 4178
Credit: 4,647,982
RAC: 0
United States
Message 323459 - Posted: 2 Jun 2006, 15:44:06 UTC

LOL, don't worry about those long run times and keep that "old timer" going! :-)

Acutually IMHO it does pretty good for itself, I had one that took more than 800 Ksecs, and that was on a NT4 box so the CPU time was correct. ;-)

Alinator
ID: 323459 · Report as offensive
Profile The MariahNet Network
Avatar

Send message
Joined: 14 Jul 99
Posts: 173
Credit: 2,469,357
RAC: 0
United States
Message 324648 - Posted: 3 Jun 2006, 6:05:46 UTC

The longer times, I can pretty much see, are basically the adjustment for CPU processing power advancement over the years. So for all the older CPUs, they can still run SETI@home stuff, but slower than before relative to the kind of CPU work of the here and now compared to the past.

I do believe long, long ago, I was able to run on a 486 without too much trouble. It then became a joke to run SETI@home on it. (I think it was SETI stuff, or it could have been something else. But I'm pretty sure it was SETI.)

Time changes, CPU processing power advances, software advances accordingly (or should), and all older things become proportionally slower and then, later on, completely obsolete. It's sad to hear, relative to what we have, at some point in time, but time moves on and technology advances. :-/

But if something still works, use it when and where it is still practical.

E
ID: 324648 · Report as offensive
Profile Blanchjoe

Send message
Joined: 27 Nov 02
Posts: 12
Credit: 7,148,533
RAC: 5
United States
Message 333128 - Posted: 10 Jun 2006, 16:35:43 UTC - in response to Message 324648.  

The longer times, I can pretty much see, are basically the adjustment for CPU processing power advancement over the years. So for all the older CPUs, they can still run SETI@home stuff, but slower than before relative to the kind of CPU work of the here and now compared to the past.

I do believe long, long ago, I was able to run on a 486 without too much trouble. It then became a joke to run SETI@home on it. (I think it was SETI stuff, or it could have been something else. But I'm pretty sure it was SETI.)

Time changes, CPU processing power advances, software advances accordingly (or should), and all older things become proportionally slower and then, later on, completely obsolete. It's sad to hear, relative to what we have, at some point in time, but time moves on and technology advances. :-/

But if something still works, use it when and where it is still practical.

E


ID: 333128 · Report as offensive
Profile Blanchjoe

Send message
Joined: 27 Nov 02
Posts: 12
Credit: 7,148,533
RAC: 5
United States
Message 333137 - Posted: 10 Jun 2006, 16:41:59 UTC - in response to Message 324648.  

The longer times, I can pretty much see, are basically the adjustment for CPU processing power advancement over the years. So for all the older CPUs, they can still run SETI@home stuff, but slower than before relative to the kind of CPU work of the here and now compared to the past.

I do believe long, long ago, I was able to run on a 486 without too much trouble. It then became a joke to run SETI@home on it. (I think it was SETI stuff, or it could have been something else. But I'm pretty sure it was SETI.)

Time changes, CPU processing power advances, software advances accordingly (or should), and all older things become proportionally slower and then, later on, completely obsolete. It's sad to hear, relative to what we have, at some point in time, but time moves on and technology advances. :-/

But if something still works, use it when and where it is still practical.

E

Hello E,

Thanks for the insight, and yes I understand your commentary all too well. I am old enought to remember a time when a machine stayed current for the better part of its functional life, however with the advent of computers a machine is viable only for a very limited perieod of time. While my 333mhz P2 is a dinasoar compared to most, it is in real terms only about ten years old, but in computer development is ancient. I use a Gateway Destination system which is acting as my television, home computer, movie player and SETI number cruncher all at once 24-7-365 on Win98 1st Version. Needless to say it has taken some bubble gum and bailing wire to keep it going over the years on a wireless DSL network, and frustrating as you can imagine, but the cost of replacing what this system does for me for something current is prohibitive at the moment. Take care and thanks! Blanchjoe

ID: 333137 · Report as offensive
Profile RePhLeX

Send message
Joined: 3 Dec 05
Posts: 76
Credit: 128,962
RAC: 0
United Kingdom
Message 333779 - Posted: 11 Jun 2006, 11:48:57 UTC

http://setiathome.berkeley.edu/results.php?hostid=2351752


180,000 secs to crunch a wu, and 30 credits..Nice.






ID: 333779 · Report as offensive
Profile Saenger
Volunteer tester
Avatar

Send message
Joined: 3 Apr 99
Posts: 2452
Credit: 33,281
RAC: 0
Germany
Message 333781 - Posted: 11 Jun 2006, 11:52:39 UTC - in response to Message 333779.  

http://setiathome.berkeley.edu/results.php?hostid=2351752


180,000 secs to crunch a wu, and 30 credits..Nice.


So what?
It's a GenuineIntel Pentium(r) II Processor, benchmarked with
Measured floating point speed 263.98 million ops/sec
Measured integer speed 455.65 million ops/sec
What do you expect from such an old machine?
Gruesse vom Saenger

For questions about Boinc look in the BOINC-Wiki
ID: 333781 · Report as offensive
1mp0£173
Volunteer tester

Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 334297 - Posted: 11 Jun 2006, 23:34:11 UTC - in response to Message 333137.  

While my 333mhz P2 is a dinasoar compared to most, it is in real terms only about ten years old, but in computer development is ancient.

Ten years is a bit over three standard eternities in this business.

(one standard eternity is equal to three years).

I haven't built it out yet, but I have a new Athlon 64 3000+ that cost me $99, including the motherboard.
ID: 334297 · Report as offensive
Zero_Crew

Send message
Joined: 7 Jun 06
Posts: 2
Credit: 13
RAC: 0
Ethiopia
Message 335097 - Posted: 12 Jun 2006, 22:08:49 UTC

90 Hours!?! I am on one right now that has been crunching for 60 hours and still has 111 hours to go!
75% of all statistics are lies
ID: 335097 · Report as offensive
Marck
Volunteer tester

Send message
Joined: 18 May 03
Posts: 33
Credit: 1,390,532
RAC: 0
Germany
Message 336614 - Posted: 14 Jun 2006, 14:44:18 UTC - in response to Message 335097.  

90 Hours!?! I am on one right now that has been crunching for 60 hours and still has 111 hours to go!

That's peanuts! My internet router box crunches a WU for SETI enhanced in the range of 14 *days*. ;-)
ID: 336614 · Report as offensive
Profile RePhLeX

Send message
Joined: 3 Dec 05
Posts: 76
Credit: 128,962
RAC: 0
United Kingdom
Message 338159 - Posted: 15 Jun 2006, 20:47:02 UTC
Last modified: 15 Jun 2006, 20:48:46 UTC

i thought 10k secs was bad, but 1M secs..thats just mental!...

1,000,000 sec + wus


ID: 338159 · Report as offensive
Profile [AF>france>pas-de-calais]symaski62
Volunteer tester

Send message
Joined: 12 Aug 05
Posts: 258
Credit: 100,548
RAC: 0
France
Message 338316 - Posted: 15 Jun 2006, 21:50:16 UTC - in response to Message 338159.  

i thought 10k secs was bad, but 1M secs..thats just mental!...

1,000,000 sec + wus



http://setiathome.berkeley.edu/workunit.php?wuid=78987918

Owner Marck
Created 11 Feb 2006 20:55:19 UTC
Total Credit 399.10
Recent average credit 4.25
CPU type CyrixInstead
6x86 2x Core/Bus Clock
Number of CPUs 1
Operating System Linux
2.4.27-3-586tsc
Memory 92.69 MB
Cache 976.56 KB
Measured floating point speed 39.06 million ops/sec
Measured integer speed 66.7 million ops/sec

Average upload rate 4.67 KB/sec
Average download rate 9.65 KB/sec
Average turnaround time 8.73 days
Maximum daily WU quota per CPU 100/day
Results 2

SETI@Home Informational message -9 result_overflow
with a general handicap of 80% and it makes much d' efforts for the community and s' expimer, thank you d' to be understanding.
ID: 338316 · Report as offensive
Profile KWSN - Chicken of Angnor
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 9 Jul 99
Posts: 1199
Credit: 6,615,780
RAC: 0
Austria
Message 338323 - Posted: 15 Jun 2006, 21:53:27 UTC - in response to Message 338316.  
Last modified: 15 Jun 2006, 21:56:38 UTC

Megasecond Madness! :o)

There really should be a stats category for the longest time taken to crunch a valid WU.

Crunch on!

90 hours are 324000 seconds, so about 1/4 of those 1.2 million...patience is a virtue.

i thought 10k secs was bad, but 1M secs..thats just mental!...

1,000,000 sec + wus



http://setiathome.berkeley.edu/workunit.php?wuid=78987918

Owner Marck
Created 11 Feb 2006 20:55:19 UTC
Total Credit 399.10
Recent average credit 4.25
CPU type CyrixInstead
6x86 2x Core/Bus Clock
Number of CPUs 1
Operating System Linux
2.4.27-3-586tsc
Memory 92.69 MB
Cache 976.56 KB
Measured floating point speed 39.06 million ops/sec
Measured integer speed 66.7 million ops/sec

Average upload rate 4.67 KB/sec
Average download rate 9.65 KB/sec
Average turnaround time 8.73 days
Maximum daily WU quota per CPU 100/day
Results 2


Donate to SETI@Home via PayPal!

Optimized SETI@Home apps + Information
ID: 338323 · Report as offensive
Ingleside
Volunteer developer

Send message
Joined: 4 Feb 03
Posts: 1546
Credit: 15,832,022
RAC: 13
Norway
Message 338330 - Posted: 15 Jun 2006, 22:02:57 UTC - in response to Message 338323.  

Megasecond Madness! :o)

There really should be a stats category for the longest time taken to crunch a valid WU.


Well, the longest deadline is 55.12 days, this equals 4762476 seconds...

Oh, and the shortest deadline is 4.34 days, this equals 375k seconds...
ID: 338330 · Report as offensive
EricVonDaniken

Send message
Joined: 17 Apr 04
Posts: 177
Credit: 67,881
RAC: 0
United States
Message 340447 - Posted: 17 Jun 2006, 16:28:29 UTC - in response to Message 338159.  
Last modified: 17 Jun 2006, 16:30:33 UTC

i thought 10k secs was bad, but 1M secs..thats just mental!...

1,000,000 sec + wus


That was a 1M sec task for their computer. For yours it would be ~10K sec task.

...and if we had SSE 1/2/3 support in the standard app for CPU's like yours that supported it, it would only be a ~2K sec task for your computer.
since yours is not the highest benching host participating, there is a good chance that w/ SSE 1/2/3 support available some hosts could crunch that task in <= 200 secs.

We =really= need
a= SSE 1/2/3 support, 64b support, the option to have a client that is "stripped" of all the graphics overhead, etc etc.
IOW, we need optimized apps.
b= a more intelligent task allocation scheme.
When expected crunching time can vary by ~4-5 orders of magnitude depending on difficulty of task combined with power of host, IMHO the variability is too high.
ID: 340447 · Report as offensive
Profile KWSN - Chicken of Angnor
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 9 Jul 99
Posts: 1199
Credit: 6,615,780
RAC: 0
Austria
Message 340460 - Posted: 17 Jun 2006, 16:47:53 UTC - in response to Message 340447.  

We =really= need
a= SSE 1/2/3 support, 64b support, the option to have a client that is "stripped" of all the graphics overhead, etc etc.
IOW, we need optimized apps.
b= a more intelligent task allocation scheme.
When expected crunching time can vary by ~4-5 orders of magnitude depending on difficulty of task combined with power of host, IMHO the variability is too high.

Not to sound like a know-it-all, but did you know that the source code is freely available?
Both for Windows and Linux.

You are welcome to compile your own client that has no graphics (it's a configure switch on Linux, similar on Windows) and/or that's optimized for your hardware platform.

You keep telling us what we need - why don't you put your expertise to work instead?

Grab a copy of the sources and hack away. GPL means you can have your cake and eat it, too.

Regards,
Simon.
Donate to SETI@Home via PayPal!

Optimized SETI@Home apps + Information
ID: 340460 · Report as offensive
EricVonDaniken

Send message
Joined: 17 Apr 04
Posts: 177
Credit: 67,881
RAC: 0
United States
Message 340482 - Posted: 17 Jun 2006, 17:09:38 UTC
Last modified: 17 Jun 2006, 17:10:26 UTC

Being able to see the solution to a problem and being able to implement it are two different skill sets.

In addition, I have =no= desire to be put through the sh*t Crunch3r went through.

Life is too short, I have other responsibilities, and most important...
...I'd be =way= too tempted to reach through my screen and the network to smack some manners and sense into Certain Parties if/when they had the gall to publicly question my skills and/or ethics when they have no frigging idea how to do the work they are criticising and could not come close to paying for it if they had to rather than it being donated.

In short, I am not as nice or conflict adverse as Crunch3r.

I'm not Stallman, but I would become =most= unpleasant under such circumstances.
So I'm going to avoid them and help out in other ways that do not risk such a confrontation or consequence. :)
ID: 340482 · Report as offensive
Profile KWSN - Chicken of Angnor
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 9 Jul 99
Posts: 1199
Credit: 6,615,780
RAC: 0
Austria
Message 340491 - Posted: 17 Jun 2006, 17:23:01 UTC

For someone who wants to smack sense and manners into someone else, you do come over strong. Might work on your tone. :o)

You seem to be able to do it, you just don't want to because of possible conflict. That's your choice, but I didn't see anyone questioning your ethics or skills (as opposed to the other way around).

Right now it's doing, not talking, that's required. The how is clear, the work involved is not trivial.

Regards,
Simon.
Donate to SETI@Home via PayPal!

Optimized SETI@Home apps + Information
ID: 340491 · Report as offensive
1 · 2 · Next

Message boards : Number crunching : 90 Hours to crunch 1 task? Help!


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.