Index  | Recent Threads  | Unanswered Threads  | Who's Active  | Guidelines  | Search
 

Quick Go »
No member browsing this thread
Thread Status: Active
Total posts in this thread: 88
Posts: 88   Pages: 9   [ Previous Page | 1 2 3 4 5 6 7 8 9 | Next Page ]
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 99811 times and has 87 replies Next Thread
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: HCC Program updated to 6.08

UBUNTU 64 8.10 (Intrepid Ibex) on a q9400 (moderate o/clock)

HCC 6.08
host ID 1028512
CPU time in hours highlight (bold)
Boinc client 6.6.33
Boincview reports average WU time to completion 20 May to 24 May for all work over a 4 days - 3190.49 sec


Result__________________________Device__Status____Sent Return _______________CPU Time_____claimed / granted


X0000083650718200702121150_ 0-- Ubuntu64 Valid 23/05/10 16:10:46 24/05/10 04:40:28 0.78 17.7 / 18.4
X0000083650717200702121150_ 0-- Ubuntu64 Valid 23/05/10 16:10:46 24/05/10 08:51:08 0.86 19.5 / 18.7
X0000083650707200702121150_ 0-- Ubuntu64 Valid 23/05/10 16:10:46 24/05/10 08:51:08 0.93 21.1 / 22.1
X0000083650438200702121154_ 0-- Ubuntu64 Valid 23/05/10 16:10:46 24/05/10 08:51:08 0.92 21.0 / 22.3
X0000083650719200702121150_ 1-- Ubuntu64 Valid 23/05/10 16:10:46 24/05/10 08:51:08 0.87 19.7 / 20.2
X0000083650720200702121150_ 0-- Ubuntu64 Valid 23/05/10 16:10:46 24/05/10 04:50:54 0.81 18.4 / 18.7
X0000083650721200702121149_ 0-- Ubuntu64 Valid 23/05/10 16:10:46 24/05/10 08:51:08 0.80 18.3 / 19.0
X0000083650722200702121149_ 1-- Ubuntu64 Valid 23/05/10 16:10:46 24/05/10 08:51:08 0.77 17.6 / 17.9
X0000083650723200702121149_ 0-- Ubuntu64 Valid 23/05/10 16:10:46 24/05/10 08:51:08 0.80 18.1 / 18.1
X0000083650725200702121149_ 0-- Ubuntu64 Valid 23/05/10 16:10:46 24/05/10 08:51:08 0.84 19.1 / 18.2

EDIT added boinc client information
----------------------------------------
[Edit 2 times, last edit by Former Member at May 24, 2010 2:11:54 PM]
[May 24, 2010 12:01:10 PM]   Link   Report threatening or abusive post: please login first  Go to top 
therealjcool
Cruncher
Joined: Feb 22, 2008
Post Count: 10
Status: Offline
Reply to this Post  Reply with Quote 
Re: HCC Program updated to 6.08

I'm doing a direct comparison between Win7 x64 and Ubuntu 10.04 as we speak. Check back later for more results, but for now it seems that Linux is indeed a lot faster on HCC.

http://www.xtremesystems.org/forums/showpost.php?p=4406365&postcount=31
----------------------------------------

[May 26, 2010 4:04:32 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Sekerob
Ace Cruncher
Joined: Jul 24, 2005
Post Count: 20043
Status: Offline
Reply to this Post  Reply with Quote 
Re: HCC Program updated to 6.08

My Q6600 system [at stock] between W7-64 and and Lucid [Generic Kernel] gave a full 2x faster under latter. If you don't achieve that, you might have to work your timings more [something I don't do] :>)

PS: the Linux is of course 64 bit, thinking it's automatically choosing the correct bit size during installation [saw no question]
----------------------------------------
WCG Global & Research > Make Proposal Help: Start Here!
Please help to make the Forums an enjoyable experience for All!
----------------------------------------
[Edit 1 times, last edit by Sekerob at May 26, 2010 4:57:03 PM]
[May 26, 2010 4:48:58 PM]   Link   Report threatening or abusive post: please login first  Go to top 
therealjcool
Cruncher
Joined: Feb 22, 2008
Post Count: 10
Status: Offline
Reply to this Post  Reply with Quote 
Re: HCC Program updated to 6.08

Work my timings? Ram timings you mean? They won't matter much on an i7, if at all.
I'm a complete Linux noob.. tried it for the first time today and am glad I got it working.

Anyway, the architectural differences between a 32nm Westmere-based Hexacore and an old 65nm Kentsfield Quad may account for different results ;)

Also, the test sample size with 15 WUs isn't that big, I know. Still, it should give us a pretty good idea.
----------------------------------------

[May 26, 2010 6:23:13 PM]   Link   Report threatening or abusive post: please login first  Go to top 
therealjcool
Cruncher
Joined: Feb 22, 2008
Post Count: 10
Status: Offline
Reply to this Post  Reply with Quote 
Re: HCC Program updated to 6.08

Sekerob (or anyone who might know it) - can you confirm or deny that the HCC server sends out the exact same work units to Windows and Linux clients?

If they do, then Linux is definitely more efficient, yet claiming at a lower rate than Windows. Here are my findings so far:

And another fresh install of Ubuntu 10.04 on a Core i7 980X (Hexacore; Frequency: 3,68Ghz; Cache: 12MB L3; Threads: 12). I want to compare runtime and claimed/granted credit for Win7 x64 vs. Ubuntu 10.04 x64.

Runtime per WU on Win7 (30-WU average): 1,876h
Claimed credit per WU on Win7 (30-WU average): 52,61
Granted credit per WU on Win7 (30-WU average): 42,82
Claimed vs. granted ratio on Win7 (30-WU average): 81,4%

Granted credits per hour (per thread): 22,83
Granted credits per hour (CPU total): 273,9
-----------------------------------------------------------

Runtime per WU on Ubuntu 10.04 (30-WU average): 1,072h
Claimed credit per WU on Ubuntu 10.04 (30-WU average): 30,71
Granted credit per WU on Ubuntu 10.04 (30-WU average): 22,55
Claimed vs. granted ratio on Ubuntu 10.04 (30-WU average): 73,4%

Credits per hour (per thread): 21,04
Credits per hour (CPU total): 252,43

So... what have we here.
From the looks of it, the Linux client is definitely more efficient. It completes the WUs a lot faster. However, the Windows client gets higher granted credit.
Now the real question is whether Linux and Windows client are getting the exact same work units.

Either way, I'm a little disappointed. I had hoped to present to everyone the "better" OS for crunching, yet it seems people have to choose between getting more science done (if the WUs are the same) or getting higher PPD.

----------------------------------------

----------------------------------------
[Edit 2 times, last edit by therealjcool at May 27, 2010 10:19:20 AM]
[May 27, 2010 10:16:23 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Randzo
Senior Cruncher
Slovakia
Joined: Jan 10, 2008
Post Count: 339
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: HCC Program updated to 6.08

Hello therealjcool HCC is very Memory intensive app not amount, but performance.
The new app is much more efective but still very memory intevsive.
So if you have hexacore with HT so 12 threads your memory is overused. And so if onother one with linux get the same WU as you he has maybe only 2 or 4 threads so his memory can handle it better. So you he finishes task earlier and he claims less credit you get less credit. You are overclaiming aren`t you?
And linux and win gets the same WU. First requester determinates the OS also for his wingman.
[May 27, 2010 11:27:48 AM]   Link   Report threatening or abusive post: please login first  Go to top 
therealjcool
Cruncher
Joined: Feb 22, 2008
Post Count: 10
Status: Offline
Reply to this Post  Reply with Quote 
Re: HCC Program updated to 6.08

Hi Radzo,

well, the hexacore may have 12 Threads, but it also has a huge bandwidth. That 980X is running 6GB of DDR3 at 1600Mhz in Triple Channel, which gives it roughly 300-400% more bandwidth than a standard Q6600 on DDR2. So I wouldn't say the i7 is slowed down by the memory.
----------------------------------------

[May 27, 2010 12:49:00 PM]   Link   Report threatening or abusive post: please login first  Go to top 
uplinger
Former World Community Grid Tech
Joined: May 23, 2005
Post Count: 3952
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: HCC Program updated to 6.08

HCC does not send out the same exact work unit to both windows and linux. It is as Randzo says, when the first work unit in the quorum is picked up, the OS is set for the wingman. This means linux compares with linux. Windows compares with windows....

A batch consists of about 1500 work units, these work units are split between all the OS's, so there is no discrimination based on batch level.

-Uplinger
[May 27, 2010 1:16:23 PM]   Link   Report threatening or abusive post: please login first  Go to top 
TimAndHedy
Senior Cruncher
Joined: Jan 27, 2009
Post Count: 267
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: HCC Program updated to 6.08

when the first work unit in the quorum is picked up, the OS is set for the wingman


Why? Aren't they comparable results?
[May 27, 2010 2:37:24 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Randzo
Senior Cruncher
Slovakia
Joined: Jan 10, 2008
Post Count: 339
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: HCC Program updated to 6.08

They are not the same but staticticaly are the same.
Of course each WU is unique.
It is the same question if the two cups of water from one well contains the same water. It is the same but it is not. Depends on point of view.
So they are comparable but you need moore observation. Statistics recomend more than 30 observations but more is better :-)
[May 27, 2010 2:50:22 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Posts: 88   Pages: 9   [ Previous Page | 1 2 3 4 5 6 7 8 9 | Next Page ]
[ Jump to Last Post ]
Post new Thread