Новые клиенты для OGR Налаштування

[Tamagoch]

Текущая серия:

==========================================================================
Version change log: $Id: CHANGES.txt,v 1.66 2009/06/03 11:29:22 andreasb Exp $
==========================================================================

2.9100
------

2.9109.518 fix: all: OGR-NG codebase fixes (#4295)
fix: all: OGR-NG incorrect node count after full restart
of combined stub (#4290)
imp: all: a kind of stress test for OGR-NG
imp: cellbe: faster OGR-NG SPU ASM core
imp: all: log errors to file during tests and buffer update
fix: x86: possible division by zero in cpuinfo
imp: x86: CPU detect and core select updates (#4297,#4299,#4301)
imp: x86: AMD Opteron/Phenom detection (#4326,#4256)
fix: all: logging issues with more then 26 crunchers (#4345)
imp: ppc: recognize more PowerPC processors (#4349)
imp: ppc: core selection for Power6 series (#4349)
fix: win64: only 32 processors were used (#4352)


2.9105-511 fix: ppc: Fixed scalar and hybrid OGR-NG cores (#4187)
imp: cellbe: detect linux kernel SPU scheduler (spufs) bugs
new: linux: runs-on-battery-power detection via sysfs

2.9104-510 imp: all: handle lack of a supported cpu/gpu more gracefully
imp: gpgpu: terminate on unsupported driver/library installations
imp: gpgpu: raise rc5-72 buffer limit to 10000
new: cuda: switch to CUDA 2.1 (Linux and Windows only)
new: cuda: bundle runtime library (libcudart.so.2/libcudart.dylib/
cudart.dll) with the released client
imp: cuda: log errors to logfile (if enabled)
new: x86: three OGR-NG SSE2 ASM cores from Craig Johnston (#4177)
imp: cellbe: faster OGR-NG SPU ASM core


Загрузить клиент -=- FAQ по настройке

-->> https://distributed.net/download/prerelease.php <<--


Windows 32bit
1 [x86/Zipped] v2.9109.518 2010-11-10 http | ftp
2 [x86/CUDA-3.1] v2.9109.518 2010-11-10 http | ftp

Windows 64bit
3 [AMD64] v2.9109.518 2010-11-10 http | ftp

Mac OS X/Darwin
4 [AMD64] v2.9109.518 2010-11-10 http | ftp

[Tamagoch]

ВЫВОД: _пока_ щитать лучше 32-х битным клиентом.

сделайте еще тест на АМД, потому что у меня:

dnetc v2.9103-509-GTR-09013001 for Win64 (WindowsNT 5.2).
[Mar 02 20:38:02 UTC] OGR-NG: Benchmark for core #0 (FLEGE-64 2.0)
0.00:00:16.57 [29,966,163 nodes/sec]

dnetc v2.9103-509-GTR-09013001 for Win32 (WindowsNT 5.2).
[Mar 03 06:00:53 UTC] OGR-NG: using core #0 (FLEGE 2.0).
[Mar 03 06:01:13 UTC] OGR-NG: Benchmark for core #0 (FLEGE 2.0)
0.00:00:16.34 [18,055,168 nodes/sec]
[Mar 03 06:01:13 UTC] OGR-NG: using core #1 (rt-asm-generic).
[Mar 03 06:01:32 UTC] OGR-NG: Benchmark for core #1 (rt-asm-generic)
0.00:00:16.62 [19,283,748 nodes/sec]
[Mar 03 06:01:32 UTC] OGR-NG: using core #2 (rt-asm-mmx).
[Mar 03 06:01:52 UTC] OGR-NG: Benchmark for core #2 (rt-asm-mmx)
0.00:00:16.40 [23,137,884 nodes/sec]

[A1ex01]

качаем новых клиентов, есть прирост на sse2

[Death]

dnetc v2.9104-510-GTR-09042813 for Win32 (WindowsNT 5.1).
Using email address (distributed.net ID) 'satan\nebesa.com'

[May 05 11:56:52 UTC] Automatic processor type detection found
an Intel Pentium 4/D/4-M/Celeron/Xeon processor.
[May 05 11:56:52 UTC] OGR-NG: using core #2 (rt-asm-mmx).
[May 05 11:56:52 UTC] OGR-NG: Loaded 27/25-5-55-41*
[May 05 11:56:52 UTC] OGR-NG: 2 packets remain in buff-in.og2
[May 05 11:56:52 UTC] OGR-NG: 0 packets are in buff-out.og2
[May 05 11:56:52 UTC] 1 cruncher has been started.
[May 05 11:56:56 UTC] OGR-NG: using core #0 (FLEGE 2.0).
[May 05 11:57:16 UTC] OGR-NG: Benchmark for core #0 (FLEGE 2.0)
0.00:00:16.51 [9,373,228 nodes/sec]
[May 05 11:57:16 UTC] OGR-NG: using core #1 (rt-asm-generic).
[May 05 11:57:35 UTC] OGR-NG: Benchmark for core #1 (rt-asm-generic)
0.00:00:16.48 [8,736,251 nodes/sec]
[May 05 11:57:35 UTC] OGR-NG: using core #2 (rt-asm-mmx).
[May 05 11:57:55 UTC] OGR-NG: Benchmark for core #2 (rt-asm-mmx)
0.00:00:17.14 [10,922,666 nodes/sec]
[May 05 11:57:55 UTC] OGR-NG: using core #3 (cj-asm-sse2).
[May 05 11:58:15 UTC] OGR-NG: Benchmark for core #3 (cj-asm-sse2)
0.00:00:17.20 [10,895,360 nodes/sec]
[May 05 11:58:15 UTC] OGR-NG: using core #4 (cj-asm-sse-p4).
[May 05 11:58:35 UTC] OGR-NG: Benchmark for core #4 (cj-asm-sse-p4)
0.00:00:16.76 [10,127,160 nodes/sec]
[May 05 11:58:35 UTC] OGR-NG: using core #5 (cj-asm-sse-k8).
[May 05 11:58:55 UTC] OGR-NG: Benchmark for core #5 (cj-asm-sse-k8)
0.00:00:17.23 [9,386,858 nodes/sec]
[May 05 11:58:55 UTC] OGR-NG benchmark summary :
Default core : #4 (cj-asm-sse-p4)
Fastest core : #2 (rt-asm-mmx)
[May 05 11:58:55 UTC] Core #2 is significantly faster than the default core.
Please file a bug report along with the output of -cpuinfo.

Реально всё равно ммх на старом селероне рулит. )) Но не сильно. Ещё дома на атлоне попробую.
Давайте кидайте результаты -benchmark.

[Water]

В связи с апгрейдом новым клиентом жалко было тормозить:
[May 05 07:24:53 UTC] OGR-NG: Summary: 4708 packets (609934.71 stats units)
69.01:36:31.89 - [102.21 Mnodes/s]

[Death]

2 месяца аптайма? ВААУ

[Water]

Со 2-го дня работы проекта
В связи с апгрейдом пришлось клиента перезапустить
Зато вместо 102 мегакейс 124.

[rrappee]

[May 05 13:50:56 UTC] 1 cruncher has been started.
[May 05 13:53:22 UTC] OGR-NG:27/31-14-30-25-16@16+10-9-4-50- [2,167,209,290]
[May 05 13:53:23 UTC] OGR-NG: using core #0 (FLEGE 2.0).
[May 05 13:53:42 UTC] OGR-NG: Benchmark for core #0 (FLEGE 2.0)
0.00:00:16.68 [15,427,495 nodes/sec]
[May 05 13:53:42 UTC] OGR-NG: using core #1 (rt-asm-generic).
[May 05 13:54:01 UTC] OGR-NG: Benchmark for core #1 (rt-asm-generic)
0.00:00:16.81 [14,773,669 nodes/sec]
[May 05 13:54:01 UTC] OGR-NG: using core #2 (rt-asm-mmx).
[May 05 13:54:21 UTC] OGR-NG: Benchmark for core #2 (rt-asm-mmx)
0.00:00:16.78 [22,557,138 nodes/sec]
[May 05 13:54:21 UTC] OGR-NG: using core #3 (cj-asm-sse2).
[May 05 13:54:40 UTC] OGR-NG: Benchmark for core #3 (cj-asm-sse2)
0.00:00:16.60 [22,679,552 nodes/sec]
[May 05 13:54:40 UTC] OGR-NG: using core #4 (cj-asm-sse-p4).
[May 05 13:55:00 UTC] OGR-NG: Benchmark for core #4 (cj-asm-sse-p4)
0.00:00:16.79 [25,175,414 nodes/sec]
[May 05 13:55:00 UTC] OGR-NG: using core #5 (cj-asm-sse-k8).
[May 05 13:55:19 UTC] OGR-NG: Benchmark for core #5 (cj-asm-sse-k8)
0.00:00:16.70 [24,231,936 nodes/sec]
[May 05 13:55:19 UTC] OGR-NG benchmark summary :
Default core : #4 (cj-asm-sse-p4)
Fastest core : #4 (cj-asm-sse-p4)

процессор селерон 346@3681, win32
так что да, прирост есть

[rrappee]

[May 05 19:00:44 UTC] OGR-NG: using core #0 (FLEGE 2.0).
[May 05 19:01:03 UTC] OGR-NG: Benchmark for core #0 (FLEGE 2.0)
0.00:00:16.21 [30,239,857 nodes/sec]
[May 05 19:01:03 UTC] OGR-NG: using core #1 (rt-asm-generic).
[May 05 19:01:22 UTC] OGR-NG: Benchmark for core #1 (rt-asm-generic)
0.00:00:16.10 [38,738,276 nodes/sec]
[May 05 19:01:22 UTC] OGR-NG: using core #2 (rt-asm-mmx).
[May 05 19:01:41 UTC] OGR-NG: Benchmark for core #2 (rt-asm-mmx)
0.00:00:16.34 [53,447,861 nodes/sec]
[May 05 19:01:41 UTC] OGR-NG: using core #3 (cj-asm-sse2).
[May 05 19:02:00 UTC] OGR-NG: Benchmark for core #3 (cj-asm-sse2)
0.00:00:16.43 [64,520,192 nodes/sec]
[May 05 19:02:00 UTC] OGR-NG: using core #4 (cj-asm-sse-p4).
[May 05 19:02:18 UTC] OGR-NG: Benchmark for core #4 (cj-asm-sse-p4)
0.00:00:16.32 [52,207,349 nodes/sec]
[May 05 19:02:18 UTC] OGR-NG: using core #5 (cj-asm-sse-k8).
[May 05 19:02:38 UTC] OGR-NG: Benchmark for core #5 (cj-asm-sse-k8)
0.00:00:16.73 [54,555,780 nodes/sec]
[May 05 19:02:38 UTC] OGR-NG benchmark summary :
Default core : #3 (cj-asm-sse2)
Fastest core : #3 (cj-asm-sse2)

а вот для сравнения Core 7200@3825. win32
однозначно лучше
спасибо за инфу

ЗЫ:Tamagoch привет из ЦРП

[Death]

athlon 2200

dnetc v2.9104-510-GTR-09042813 for Win32 (WindowsNT 5.1).

[May 05 20:11:50 UTC] Automatic processor type detection found an AMD Athlon XP/MP or Sempron (Thoroughbred) processor.
[May 05 20:11:50 UTC] OGR-NG: using core #0 (FLEGE 2.0).
[May 05 20:12:14 UTC] OGR-NG: Benchmark for core #0 (FLEGE 2.0)
0.00:00:19.98 [13,686,847 nodes/sec]
[May 05 20:12:14 UTC] OGR-NG: using core #1 (rt-asm-generic).
[May 05 20:12:34 UTC] OGR-NG: Benchmark for core #1 (rt-asm-generic)
0.00:00:17.26 [13,973,839 nodes/sec]
[May 05 20:12:34 UTC] OGR-NG: using core #2 (rt-asm-mmx).
[May 05 20:12:56 UTC] OGR-NG: Benchmark for core #2 (rt-asm-mmx)
0.00:00:17.95 [17,565,356 nodes/sec]
[May 05 20:12:56 UTC] OGR-NG: using core #4 (cj-asm-sse-p4).
[May 05 20:13:16 UTC] OGR-NG: Benchmark for core #4 (cj-asm-sse-p4)
0.00:00:17.04 [16,928,251 nodes/sec]
[May 05 20:13:16 UTC] OGR-NG: using core #5 (cj-asm-sse-k8).
[May 05 20:13:36 UTC] OGR-NG: Benchmark for core #5 (cj-asm-sse-k8)
0.00:00:16.48 [18,330,523 nodes/sec]
[May 05 20:13:36 UTC] OGR-NG benchmark summary :
Default core : #5 (cj-asm-sse-k8)
Fastest core : #5 (cj-asm-sse-k8)

прирост понтовый. далеко не 20%. так шо селероны и атлоны выиграли немного. точнее селерон ваще ни фига не выиграл. на нём ммх рулит по прежнему.

вот -цпуинфо атлона.

dnetc v2.9104-510-GTR-09042813 for Win32 (WindowsNT 5.1).

Automatic processor identification tag: 40006081
name: AMD Athlon XP/MP or Sempron (Thoroughbred)
features: MMX AMD_MMX+ 3DNOW 3DNOW+ SSE
Estimated processor clock speed (0 if unknown): 2100 MHz
Number of processors detected by this client: 1
Number of processors supported by this client: 128

Raw processor informations :
Vendor ID : "AuthenticAMD"
Brand ID : "AMD Sempron™ "
F_00000000 : 00000001 68747541 444D4163 69746E65
F_00000001 : 00000681 00000000 00000000 0383FBFF
F_80000000 : 80000008 68747541 444D4163 69746E65
F_80000001 : 00000781 00000000 00000000 C1CBFBFF
F_80000002 : 20444D41 706D6553 286E6F72 20296D74
F_80000003 : 00000020 00000000 00000000 00000000
F_80000004 : 00000000 00000000 00000000 00000000
F_80000005 : 0408FF08 FF20FF10 40020140 40020140
F_80000006 : 00000000 41004100 01008140 00000000
F_80000007 : 00000000 00000000 00000000 00000001
F_80000008 : 00002022 00000000 00000000 00000000

[Tamagoch]

ЗЫ:Tamagoch привет из ЦРП

привет

прирост понтовый. далеко не 20%. так шо селероны и атлоны выиграли немного. точнее селерон ваще ни фига не выиграл. на нём ммх рулит по прежнему.

ну кое-как этот прирост на пару лет сократит продолжительность проекта

[Death]

Tamagoch, шо то я твоих бенчей не вижу ))))

[A1ex01]

CPU core2-6300@2.5GHz

dnetc v2.9104-510-GTR-09042813 for Win32 (WindowsNT 6.0).

[May 06 10:26:24 UTC] OGR-NG: using core #0 (FLEGE 2.0).
[May 06 10:26:44 UTC] OGR-NG: Benchmark for core #0 (FLEGE 2.0)
0.00:00:17.02 [19,548,828 nodes/sec]
[May 06 10:26:44 UTC] OGR-NG: using core #1 (rt-asm-generic).
[May 06 10:27:03 UTC] OGR-NG: Benchmark for core #1 (rt-asm-generic)
0.00:00:16.14 [24,945,127 nodes/sec]
[May 06 10:27:03 UTC] OGR-NG: using core #2 (rt-asm-mmx).
[May 06 10:27:22 UTC] OGR-NG: Benchmark for core #2 (rt-asm-mmx)
0.00:00:17.06 [35,294,376 nodes/sec]
[May 06 10:27:22 UTC] OGR-NG: using core #3 (cj-asm-sse2).
[May 06 10:27:41 UTC] OGR-NG: Benchmark for core #3 (cj-asm-sse2)
0.00:00:16.38 [42,568,849 nodes/sec]
[May 06 10:27:41 UTC] OGR-NG: using core #4 (cj-asm-sse-p4).
[May 06 10:28:02 UTC] OGR-NG: Benchmark for core #4 (cj-asm-sse-p4)
0.00:00:17.16 [34,112,753 nodes/sec]
[May 06 10:28:02 UTC] OGR-NG: using core #5 (cj-asm-sse-k8).
[May 06 10:28:20 UTC] OGR-NG: Benchmark for core #5 (cj-asm-sse-k8)
0.00:00:16.14 [34,079,806 nodes/sec]
[May 06 10:28:20 UTC] OGR-NG benchmark summary :
Default core : #3 (cj-asm-sse2)
Fastest core : #3 (cj-asm-sse2)

~+20%

[rrappee]

[Jan 27 21:01:02 UTC] OGR-NG: using core #0 (FLEGE 2.0).
[Jan 27 21:01:21 UTC] OGR-NG: Benchmark for core #0 (FLEGE 2.0)
0.00:00:16.73 [15,731,348 nodes/sec]
[Jan 27 21:01:21 UTC] OGR-NG: using core #1 (rt-asm-generic).
[Jan 27 21:01:40 UTC] OGR-NG: Benchmark for core #1 (rt-asm-generic)
0.00:00:16.67 [16,516,129 nodes/sec]
[Jan 27 21:01:40 UTC] OGR-NG: using core #2 (rt-asm-mmx).
[Jan 27 21:02:00 UTC] OGR-NG: Benchmark for core #2 (rt-asm-mmx)
0.00:00:17.25 [21,003,312 nodes/sec]
[Jan 27 21:02:00 UTC] OGR-NG: using core #4 (cj-asm-sse-p4).
[Jan 27 21:02:21 UTC] OGR-NG: Benchmark for core #4 (cj-asm-sse-p4)
0.00:00:17.43 [19,968,312 nodes/sec]
[Jan 27 21:02:21 UTC] OGR-NG: using core #5 (cj-asm-sse-k8).
[Jan 27 21:02:41 UTC] OGR-NG: Benchmark for core #5 (cj-asm-sse-k8)
0.00:00:16.82 [21,845,333 nodes/sec]
[Jan 27 21:02:41 UTC] OGR-NG benchmark summary :
Default core : #5 (cj-asm-sse-k8)
Fastest core : #5 (cj-asm-sse-k8)

процессор AMD k7 Barton@2425, win32

[rrappee]

[May 06 14:43:37 UTC] OGR-NG: using core #0 (FLEGE 2.0).
[May 06 14:43:57 UTC] OGR-NG: Benchmark for core #0 (FLEGE 2.0)
0.00:00:16.36 [17,054,051 nodes/sec]
[May 06 14:43:57 UTC] OGR-NG: using core #1 (rt-asm-generic).
[May 06 14:44:15 UTC] OGR-NG: Benchmark for core #1 (rt-asm-generic)
0.00:00:16.32 [16,211,989 nodes/sec]
[May 06 14:44:15 UTC] OGR-NG: using core #2 (rt-asm-mmx).
[May 06 14:44:35 UTC] OGR-NG: Benchmark for core #2 (rt-asm-mmx)
0.00:00:16.75 [24,892,467 nodes/sec]
[May 06 14:44:35 UTC] OGR-NG: using core #3 (cj-asm-sse2).
[May 06 14:44:54 UTC] OGR-NG: Benchmark for core #3 (cj-asm-sse2)
0.00:00:16.76 [25,473,542 nodes/sec]
[May 06 14:44:54 UTC] OGR-NG: using core #4 (cj-asm-sse-p4).
[May 06 14:45:13 UTC] OGR-NG: Benchmark for core #4 (cj-asm-sse-p4)
0.00:00:16.54 [27,567,596 nodes/sec]
[May 06 14:45:13 UTC] OGR-NG: using core #5 (cj-asm-sse-k8).
[May 06 14:45:32 UTC] OGR-NG: Benchmark for core #5 (cj-asm-sse-k8)
0.00:00:16.59 [26,881,050 nodes/sec]
[May 06 14:45:33 UTC] OGR-NG benchmark summary :
Default core : #4 (cj-asm-sse-p4)
Fastest core : #4 (cj-asm-sse-p4)

процессор Pentium D(presler)@4062 & win32

[A1ex01]

xeon 5110@1.6GHz

[May 06 14:44:14 UTC] OGR-NG: using core #0 (FLEGE 2.0).
[May 06 14:44:35 UTC] OGR-NG: Benchmark for core #0 (FLEGE 2.0)
0.00:00:17.21 [13,271,769 nodes/sec]
[May 06 14:44:35 UTC] OGR-NG: using core #1 (rt-asm-generic).
[May 06 14:44:55 UTC] OGR-NG: Benchmark for core #1 (rt-asm-generic)
0.00:00:17.17 [16,359,983 nodes/sec]
[May 06 14:44:55 UTC] OGR-NG: using core #2 (rt-asm-mmx).
[May 06 14:45:15 UTC] OGR-NG: Benchmark for core #2 (rt-asm-mmx)
0.00:00:17.31 [22,912,671 nodes/sec]
[May 06 14:45:15 UTC] OGR-NG: using core #3 (cj-asm-sse2).
[May 06 14:45:36 UTC] OGR-NG: Benchmark for core #3 (cj-asm-sse2)
0.00:00:17.93 [30,357,587 nodes/sec]
[May 06 14:45:36 UTC] OGR-NG: using core #4 (cj-asm-sse-p4).
[May 06 14:45:57 UTC] OGR-NG: Benchmark for core #4 (cj-asm-sse-p4)
0.00:00:18.28 [23,972,637 nodes/sec]
[May 06 14:45:57 UTC] OGR-NG: using core #5 (cj-asm-sse-k8).
[May 06 14:46:19 UTC] OGR-NG: Benchmark for core #5 (cj-asm-sse-k8)
0.00:00:17.78 [23,271,354 nodes/sec]
[May 06 14:46:19 UTC] OGR-NG benchmark summary :
Default core : #3 (cj-asm-sse2)
Fastest core : #3 (cj-asm-sse2)

~+30%