FAH 硬體及客戶端綜合討論 (part 2)

oegg

榮譽會員
已加入
3/13/04
訊息
1,389
互動分數
0
點數
36
2684 讓 6C/12T 4.2G -smp 11 TPF 大約31分55秒
2684 讓 6C/12T 4.2G -smp 12 TPF 大約31分10秒

-smp 12 是比 -smp 11 快一點但不明顯
如果還要跑 GPU 就還是 用 -smp 11
 

kazama

進階會員
已加入
2/11/10
訊息
218
互動分數
0
點數
16
昨天晚上將一台i7-920平台的作業系統由Linux更換為Windows XP 32bit,
運算project 2684 bigadv smp -8以工作管理員檢視系統總使用RAM量
為1.5GB。
 

OverFold

高級會員
已加入
7/29/09
訊息
541
互動分數
20
點數
18
小弟也開始我的2684工作生涯,以i7-930@4GHz運算頭10個%計平均TPF約需
44:20(約值21000PPD),新的A3 Core Fullloading溫度確實比上一個版本較為
涼快.....最高約93度(以前A2頻頻緊迫著100度),Windows 7 64-bit下跑那個
FAHSMP629 64-bit Console版Client記憶體使用量不過1GB,總記憶體用量亦
未過1.9GB,話雖如此,系統若非專為作F@H-SMP而還有其他大大小小用途的話
,保險些還是建議提供基本的3-4GB RAM較為穩妥。
 

ShinnShinn

進階會員
已加入
2/25/09
訊息
253
互動分數
0
點數
0
http://en.fah-addict.net/news/news-0-258+bigadv-suspended-under-linux-windows-tests-begin.php

"BigAdv" suspended under Linux; Windows tests begin

As a result of a bug affecting A3-core BigAdv units on Linux, Peter Kasson has suspended the assignment of these units to Linux machines.

The cause of this bug remains unknown, but pending its resolution, you should fall back on non-BigAdv units for the time being.

If you have one handy, you can still try your hand at folding BigAdv units under a Windows installation, as this capability has just been activated for the OS. As with the Linux client, running it with the -smp and -bigadv flags on a version 6.29 client or higher will allow you to obtain BigAdv units.

The current stock of BigAdv units will remain fairly limited until new projects are started with corrected cores. In the event of a shortage, your Windows machine should fall back on A3 projects.
 

OverFold

高級會員
已加入
7/29/09
訊息
541
互動分數
20
點數
18
BAD...News....

花了近30小時運算Windows SMP版BigAdv後的結果.....

[16:56:20] Completed 95000 out of 250000 steps (38%)
[17:40:48] Completed 97500 out of 250000 steps (39%)
[18:15:03] - Autosending finished units... [July 4 18:15:03 UTC]
[18:15:03] Trying to send all finished work units
[18:15:03] + No unsent completed units remaining.
[18:15:03] - Autosend completed
[18:25:18] Completed 100000 out of 250000 steps (40%)
[00:15:03] - Autosending finished units... [July 5 00:15:03 UTC] (*****為何又再出現Auto Send?*****)
[00:15:03] Trying to send all finished work units
[00:15:03] + No unsent completed units remaining.
[00:15:03] - Autosend completed
[00:45:37] CoreStatus = C0000005 (-1073741819) (*****半小時後*****)
[00:45:37] Client-core communications error: ERROR 0xc0000005
[00:45:37] Deleting current work unit & continuing... (*****惡夢再臨*****)
[00:45:55] Trying to send all finished work units
[00:45:55] + No unsent completed units remaining.
[00:45:55] - Preparing to get new work unit...
[00:45:55] Cleaning up work directory
[00:45:55] + Attempting to get work packet
[00:45:55] Passkey found
[00:45:55] - Will indicate memory of 4094 MB
[00:45:55] - Detect CPU. Vendor: GenuineIntel, Family: 6, Model: 10, Stepping: 5

[00:45:55] - Connecting to assignment server
[00:45:55] Connecting to http://assign.stanford.edu:8080/
[00:45:56] Posted data.
[00:45:56] Initial: 43AB; - Successful: assigned to (171.67.108.22).
[00:45:56] + News From Folding@Home: Welcome to Folding@Home
[00:45:56] Loaded queue successfully.
[00:45:56] Connecting to http://171.67.108.22:8080/
[00:46:06] Posted data.
[00:46:06] Initial: 0000; - Receiving payload (expected size: 24824179)
[00:47:05] - Downloaded at ~410 kB/s
[00:47:05] - Averaged speed for that direction ~238 kB/s
[00:47:05] + Received work.
[00:47:05] + Closed connections
[00:47:10]
[00:47:10] + Processing work unit
[00:47:10] Core required: FahCore_a3.exe
[00:47:10] Core found.
[00:47:10] Working on queue slot 02 [July 5 00:47:10 UTC]
[00:47:10] + Working ...
[00:47:10] - Calling '.\FahCore_a3.exe -dir work/ -nice 19 -suffix 02 -np 8 -che
ckpoint 15 -verbose -lifeline 328 -version 629'

[00:47:10]
[00:47:10] *------------------------------*
[00:47:10] Folding@Home Gromacs SMP Core
[00:47:10] Version 2.22 (Mar 12, 2010)
[00:47:10]
[00:47:10] Preparing to commence simulation
[00:47:10] - Looking at optimizations...
[00:47:10] - Created dyn
[00:47:10] - Files status OK
[00:47:14] - Expanded 24823667 -> 30791309 (decompressed 124.0 percent)
[00:47:14] Called DecompressByteArray: compressed_data_size=24823667 data_size=3
0791309, decompressed_data_size=30791309 diff=0
[00:47:15] - Digital signature verified
[00:47:15]
[00:47:15] Project: 2684 (Run 9, Clone 14, Gen 4)
[00:47:15]
[00:47:15] Assembly optimizations on if available.
[00:47:15] Entering M.D.
[00:47:25] Completed 0 out of 250000 steps (0%)


放棄了~
還是回歸Native Linux 跑 -smp 8 算了!
Windows SMP 中臨出事前 CoreTemp 顯示i7-930其中一個core曾經紀錄最高溫度為96度
而當時剛巧Harddisk又Downloading something可能....
 

despot

進階會員
已加入
9/20/03
訊息
482
互動分數
0
點數
0
年齡
41
請問一下要下什麼參數才能強制跑bigadv
我最近怎麼一直收到普通SMP的包
話說這幾天真是熱到爆炸..
剛看紀錄假日沒開冷氣最高溫飆到100度了..瀕臨當機邊緣= =
 

kazama

進階會員
已加入
2/11/10
訊息
218
互動分數
0
點數
16
請問一下要下什麼參數才能強制跑bigadv
我最近怎麼一直收到普通SMP的包
目前native linux client暫時沒有bigadv可供運算,須以Windows SMP2 client下-bigadv
參數運算A3 bigadv。

請留意Windows SMP2 client安裝時有一個WU size選項一定要選big (>10M)才能運算
bigadv,否則只會收到一般的A3 WU。
 

kazama

進階會員
已加入
2/11/10
訊息
218
互動分數
0
點數
16
放棄了~
還是回歸Native Linux 跑 -smp 8 算了!
Windows SMP 中臨出事前 CoreTemp 顯示i7-930其中一個core曾經紀錄最高溫度為96度
而當時剛巧Harddisk又Downloading something可能....
由於目前native linux client暫時沒有bigadv可供運算,建議先試著降低oc幅度再觀察
Windows SMP2的穩定性。

昨天順利完成了兩個windows SMP2 bigadv,i7-860 oc 3.5G,core溫度均控制在
75度以下。
 
▌延伸閱讀