* Benchmark results
@ 2008-08-03 16:43 bbmailing
0 siblings, 0 replies; 16+ messages in thread
From: bbmailing @ 2008-08-03 16:43 UTC (permalink / raw)
To: xen-devel; +Cc: xen-users
[-- Attachment #1: Type: text/plain, Size: 1790 bytes --]
Hey everyone,
I hope you guys can help me explaining a pretty strange result that came up when benchmarking a native Red Hat installation against a Xen DomU (installed in a file container). I used lmbench3 on a dualcore AMD-machine. To make the results comparable, I "unplugged" all CPU cores but one in the native installation (setting the "online"-entry to 0). The DomU had only one vCPU assigned. The installed packages on both machines were identical as well. After doing the initial configuration of lmbench3 I ran a script doing 10 times a "make rerun".
All results seem to be OK somehow. But I cannot explain why Xen is faster than the native installation when creating and deleting files? Its 12ms (DomU) against 24ms (native) for creating a 0K file and 50ms (DomU) against 60ms (native) for creating a 10K file. Please find all the results attached.
Is there any way this is possible? Maybe the communication methods used by the shared device drivers make the virtual machine think the operation has been finished successfully faster than the native way? Is there is anyone out there who can help me understanding this result?
Furthermore: Is there a special recommendation for benchmarking Xen against native installations? I liked lmbench so far since it puts the focus on basic operations... what are your favourites? Is lmbench from a design perspective the wrong tool to use?
I have to find some reliable results so I am really thankful for every single hint here... please feel free to email me directly as well!
Thank you so much in advance!
Best regards,
Bjoern
____________________________________________________________________
Ihre Messenger, Communities und E-Mails jetzt in einem Programm!
WEB.DE MultiMessenger http://www.produkte.web.de/messenger/?did=3071
[-- Attachment #2: readable.txt --]
[-- Type: text/plain, Size: 17504 bytes --]
cd results && make summary percent 2>/dev/null | more
make[1]: Entering directory `/benchmark/lmbench3/results'
L M B E N C H 3 . 0 S U M M A R Y
------------------------------------
(Alpha software, do not distribute)
Basic system parameters
------------------------------------------------------------------------------
Host OS Description Mhz tlb cache mem scal
pages line par load
bytes
--------- ------------- ----------------------- ---- ----- ----- ------ ----
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.2800 1
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.0300 1
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.0200 1
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.0200 1
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.1300 1
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.0200 1
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.1400 1
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.0100 1
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.1500 1
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.0100 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0100 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0100 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0200 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0100 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0100 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0100 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0100 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0100 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0100 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0100 1
Processor, Processes - times in microseconds - smaller is better
------------------------------------------------------------------------------
Host OS Mhz null null open slct sig sig fork exec sh
call I/O stat clos TCP inst hndl proc proc proc
--------- ------------- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ----
xen-1cpu Linux 2.6.18- 2800 0.87 1.29 2.69 5.45 5.20 1.08 2.78 418. 1043 3264
xen-1cpu Linux 2.6.18- 2800 0.85 1.29 2.64 5.53 5.25 1.03 2.71 467. 1092 3159
xen-1cpu Linux 2.6.18- 2800 0.85 1.27 2.68 5.58 5.18 1.04 2.74 465. 1103 3121
xen-1cpu Linux 2.6.18- 2800 0.82 1.27 2.65 5.52 5.22 1.03 2.69 461. 1102 3146
xen-1cpu Linux 2.6.18- 2800 0.84 1.28 2.64 5.69 5.24 1.05 2.79 456. 1090 3119
xen-1cpu Linux 2.6.18- 2800 0.84 1.28 2.64 5.53 5.20 1.02 2.72 455. 1094 3125
xen-1cpu Linux 2.6.18- 2800 0.85 1.29 2.62 5.50 5.22 1.01 2.80 464. 1089 3147
xen-1cpu Linux 2.6.18- 2800 0.84 1.31 2.65 5.50 5.22 1.00 2.72 463. 1095 3130
xen-1cpu Linux 2.6.18- 2800 0.84 1.29 2.69 5.51 5.20 1.04 2.74 457. 1093 3138
xen-1cpu Linux 2.6.18- 2800 0.84 1.29 2.66 5.35 5.21 1.03 2.78 469. 1095 3133
nativ-1cp Linux 2.6.18- 2800 0.19 0.44 2.02 3.45 6.92 0.28 1.25 125. 407. 1905
nativ-1cp Linux 2.6.18- 2800 0.19 0.51 2.19 3.57 10.7 0.28 1.25 126. 404. 1903
nativ-1cp Linux 2.6.18- 2800 0.19 0.49 1.84 3.12 10.6 0.28 1.40 129. 398. 1903
nativ-1cp Linux 2.6.18- 2800 0.19 0.46 2.03 3.31 4.80 0.28 1.32 125. 403. 1910
nativ-1cp Linux 2.6.18- 2800 0.19 0.50 2.04 3.35 9.60 0.28 1.18 123. 403. 1905
nativ-1cp Linux 2.6.18- 2800 0.19 0.47 1.96 3.29 4.57 0.28 1.25 126. 399. 1894
nativ-1cp Linux 2.6.18- 2800 0.19 0.45 1.95 3.22 7.39 0.28 1.26 123. 398. 1925
nativ-1cp Linux 2.6.18- 2800 0.19 0.45 2.16 3.28 4.56 0.28 1.37 129. 400. 1919
nativ-1cp Linux 2.6.18- 2800 0.19 0.43 1.92 3.10 9.63 0.28 1.34 128. 413. 1954
nativ-1cp Linux 2.6.18- 2800 0.24 0.47 1.84 3.33 4.49 0.33 1.49 125. 410. 1931
Basic integer operations - times in nanoseconds - smaller is better
-------------------------------------------------------------------
Host OS intgr intgr intgr intgr intgr
bit add mul div mod
--------- ------------- ------ ------ ------ ------ ------
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.1
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.1
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.1
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.1
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
Basic float operations - times in nanoseconds - smaller is better
-----------------------------------------------------------------
Host OS float float float float
add mul div bogo
--------- ------------- ------ ------ ------ ------
xen-1cpu Linux 2.6.18- 1.4400 1.4500 6.3500 5.5900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 6.3500 5.5900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 6.3300 5.5900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 6.3200 5.5900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 6.3500 5.5900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 6.3200 5.6200
xen-1cpu Linux 2.6.18- 1.4300 1.4500 6.3500 5.5900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 6.3300 5.5900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 6.3500 5.5900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 6.3500 5.5900
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3500 5.6500
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3400 5.6700
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3500 5.6200
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3500 5.6400
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3500 5.6200
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3700 5.6500
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3500 5.6300
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3500 5.6600
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3500 5.6300
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3600 5.6600
Basic double operations - times in nanoseconds - smaller is better
------------------------------------------------------------------
Host OS double double double double
add mul div bogo
--------- ------------- ------ ------ ------ ------
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7600 7.2900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7800 7.2900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7700 7.2800
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7700 7.2800
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7600 7.2900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7800 7.2800
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7800 7.2800
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7600 7.2800
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7800 7.2800
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7900 7.3100
nativ-1cp Linux 2.6.18- 1.4400 1.4500 7.7800 7.3000
nativ-1cp Linux 2.6.18- 1.4300 1.4500 7.7800 7.3000
nativ-1cp Linux 2.6.18- 1.4300 1.4500 7.7800 7.3000
nativ-1cp Linux 2.6.18- 1.4300 1.4500 7.7800 7.3000
nativ-1cp Linux 2.6.18- 1.4300 1.4500 7.7800 7.3000
nativ-1cp Linux 2.6.18- 1.4300 1.4500 7.7900 7.3000
nativ-1cp Linux 2.6.18- 1.4300 1.4500 7.8200 7.3000
nativ-1cp Linux 2.6.18- 1.4300 1.4500 7.7800 7.3000
nativ-1cp Linux 2.6.18- 1.4300 1.4500 7.7800 7.3000
nativ-1cp Linux 2.6.18- 1.4300 1.4500 7.7900 7.3200
Context switching - times in microseconds - smaller is better
-------------------------------------------------------------------------
Host OS 2p/0K 2p/16K 2p/64K 8p/16K 8p/64K 16p/16K 16p/64K
ctxsw ctxsw ctxsw ctxsw ctxsw ctxsw ctxsw
--------- ------------- ------ ------ ------ ------ ------ ------- -------
xen-1cpu Linux 2.6.18- 2.4900 2.7300 5.2100 4.2200 7.2100 4.63000 38.6
xen-1cpu Linux 2.6.18- 2.5800 2.7300 5.2600 4.1400 8.0800 4.90000 47.2
xen-1cpu Linux 2.6.18- 2.5800 2.6400 5.2400 4.1100 7.5000 4.09000 46.4
xen-1cpu Linux 2.6.18- 2.7100 2.7800 5.2500 4.2000 8.0400 5.10000 43.5
xen-1cpu Linux 2.6.18- 2.5400 2.7500 5.3900 4.1100 7.6700 4.84000 44.0
xen-1cpu Linux 2.6.18- 2.5200 2.7400 5.1500 4.2100 6.7300 4.60000 40.1
xen-1cpu Linux 2.6.18- 2.6100 2.6600 5.3000 4.1800 7.5800 4.68000 41.5
xen-1cpu Linux 2.6.18- 2.5900 2.7400 5.2100 4.2000 6.8300 4.65000 43.0
xen-1cpu Linux 2.6.18- 2.5100 2.6300 5.2100 4.0400 6.6800 4.74000 43.8
xen-1cpu Linux 2.6.18- 2.5700 2.8100 5.1000 4.2200 8.5200 4.81000 43.2
nativ-1cp Linux 2.6.18- 0.7400 0.8200 3.2200 1.7600 4.1600 1.91000 37.3
nativ-1cp Linux 2.6.18- 0.6000 0.8600 3.0400 1.8100 4.8200 1.78000 38.4
nativ-1cp Linux 2.6.18- 0.6500 0.7600 3.2500 1.8200 3.8800 1.82000 40.3
nativ-1cp Linux 2.6.18- 0.6200 0.8000 3.3900 1.7700 7.1800 2.28000 38.8
nativ-1cp Linux 2.6.18- 0.7800 0.8500 3.3700 1.8600 3.7200 1.97000 36.4
nativ-1cp Linux 2.6.18- 0.6400 0.8100 3.3600 1.8100 4.3400 2.46000 36.0
nativ-1cp Linux 2.6.18- 0.5600 0.7500 3.0700 1.7400 4.1200 2.23000 33.0
nativ-1cp Linux 2.6.18- 0.6500 0.9100 3.3100 1.8700 4.3100 1.90000 37.5
nativ-1cp Linux 2.6.18- 0.6100 0.7900 3.4100 1.7200 4.5600 1.86000 33.8
nativ-1cp Linux 2.6.18- 0.5800 0.8300 3.1500 1.8200 4.1600 2.28000 37.9
*Local* Communication latencies in microseconds - smaller is better
---------------------------------------------------------------------
Host OS 2p/0K Pipe AF UDP RPC/ TCP RPC/ TCP
ctxsw UNIX UDP TCP conn
--------- ------------- ----- ----- ---- ----- ----- ----- ----- ----
xen-1cpu Linux 2.6.18- 2.490 12.1 19.5 17.6 23.7 20.8 29.2 54.
xen-1cpu Linux 2.6.18- 2.580 12.0 19.8 17.4 23.7 20.9 28.8 55.
xen-1cpu Linux 2.6.18- 2.580 12.0 19.2 17.4 23.6 21.0 28.5 55.
xen-1cpu Linux 2.6.18- 2.710 12.1 19.3 17.4 23.6 21.1 28.3 55.
xen-1cpu Linux 2.6.18- 2.540 12.9 19.3 17.4 23.7 21.2 28.7 55.
xen-1cpu Linux 2.6.18- 2.520 12.2 19.4 17.4 23.6 21.1 28.4 55.
xen-1cpu Linux 2.6.18- 2.610 12.9 19.4 17.4 23.7 21.2 28.8 55.
xen-1cpu Linux 2.6.18- 2.590 13.0 19.5 17.6 23.7 21.1 28.7 55.
xen-1cpu Linux 2.6.18- 2.510 12.0 19.3 17.2 23.8 21.3 28.6 55.
xen-1cpu Linux 2.6.18- 2.570 13.0 19.2 17.5 23.9 21.1 28.6 55.
nativ-1cp Linux 2.6.18- 0.740 4.631 7.59 9.379 12.8 12.1 16.7 126.
nativ-1cp Linux 2.6.18- 0.600 4.756 8.19 9.929 12.9 12.4 17.2 131.
nativ-1cp Linux 2.6.18- 0.650 4.589 7.62 9.214 12.7 12.2 16.7 123.
nativ-1cp Linux 2.6.18- 0.620 4.587 7.97 9.319 13.1 12.1 17.7 126.
nativ-1cp Linux 2.6.18- 0.780 4.582 8.30 10.0 13.1 12.6 16.8 128.
nativ-1cp Linux 2.6.18- 0.640 4.628 7.88 9.579 13.9 12.3 17.7 130.
nativ-1cp Linux 2.6.18- 0.560 4.575 7.80 9.656 13.2 12.5 17.5 125.
nativ-1cp Linux 2.6.18- 0.650 4.986 8.12 9.576 12.9 12.3 17.3 127.
nativ-1cp Linux 2.6.18- 0.610 4.635 7.69 10.0 13.1 12.2 16.7 125.
nativ-1cp Linux 2.6.18- 0.580 4.782 7.80 9.993 13.8 12.1 17.7 126.
File & VM system latencies in microseconds - smaller is better
-------------------------------------------------------------------------------
Host OS 0K File 10K File Mmap Prot Page 100fd
Create Delete Create Delete Latency Fault Fault selct
--------- ------------- ------ ------ ------ ------ ------- ----- ------- -----
xen-1cpu Linux 2.6.18- 11.9 9.2752 51.7 18.5 14.3K 0.848 3.97750 2.628
xen-1cpu Linux 2.6.18- 12.3 9.3171 53.8 19.4 15.4K 0.864 4.06550 2.619
xen-1cpu Linux 2.6.18- 12.3 9.6721 53.7 19.1 15.5K 0.910 4.16050 2.596
xen-1cpu Linux 2.6.18- 12.1 9.5280 41.9 19.1 15.7K 0.944 4.19680 2.588
xen-1cpu Linux 2.6.18- 12.1 9.4666 46.8 19.8 15.8K 0.892 4.18000 2.638
xen-1cpu Linux 2.6.18- 12.1 9.4604 53.5 19.7 16.5K 0.918 4.22680 2.598
xen-1cpu Linux 2.6.18- 12.3 9.7807 48.2 19.6 16.0K 0.905 4.25190 2.625
xen-1cpu Linux 2.6.18- 12.2 9.7652 52.0 20.4 16.0K 0.896 4.23290 2.695
xen-1cpu Linux 2.6.18- 12.3 9.6259 53.6 20.1 16.0K 0.867 4.26970 2.629
xen-1cpu Linux 2.6.18- 12.2 9.6403 52.7 19.6 16.1K 0.861 4.27980 2.613
nativ-1cp Linux 2.6.18- 24.8 14.5 59.5 26.4 5090.0 0.554 1.52850 4.263
nativ-1cp Linux 2.6.18- 26.3 16.1 58.7 27.3 5118.0 0.585 1.71360 4.251
nativ-1cp Linux 2.6.18- 24.9 15.4 60.5 26.3 4892.0 0.612 1.52190 1.942
nativ-1cp Linux 2.6.18- 24.4 16.1 58.7 27.2 5277.0 0.373 1.66030 1.953
nativ-1cp Linux 2.6.18- 24.3 14.7 59.5 26.5 5130.0 0.627 1.54020 1.924
nativ-1cp Linux 2.6.18- 23.9 16.1 57.8 26.1 5080.0 0.571 1.55070 1.980
nativ-1cp Linux 2.6.18- 25.6 15.5 68.7 26.8 5214.0 0.502 1.75650 1.941
nativ-1cp Linux 2.6.18- 24.7 14.4 58.3 25.4 5014.0 0.552 1.73150 1.965
nativ-1cp Linux 2.6.18- 25.6 15.2 60.4 26.4 5080.0 0.321 1.67840 1.973
nativ-1cp Linux 2.6.18- 25.8 15.2 61.6 27.0 5221.0 0.694 1.71090 1.967
*Local* Communication bandwidths in MB/s - bigger is better
-----------------------------------------------------------------------------
Host OS Pipe AF TCP File Mmap Bcopy Bcopy Mem Mem
UNIX reread reread (libc) (hand) read write
--------- ------------- ---- ---- ---- ------ ------ ------ ------ ---- -----
xen-1cpu Linux 2.6.18- 1764 2004 812. 1285.7 1889.5 834.1 819.5 1644 1121.
xen-1cpu Linux 2.6.18- 1762 1976 797. 1120.6 1405.6 705.4 724.1 1436 966.3
xen-1cpu Linux 2.6.18- 1772 1997 852. 1119.6 1420.5 718.9 727.7 1446 987.7
xen-1cpu Linux 2.6.18- 1758 2004 828. 1115.8 1429.3 720.6 730.2 1448 991.7
xen-1cpu Linux 2.6.18- 1749 1946 827. 1109.9 1434.8 721.6 733.2 1457 996.8
xen-1cpu Linux 2.6.18- 1777 1987 818. 1113.4 1440.1 725.4 742.8 1462 1001.
xen-1cpu Linux 2.6.18- 1761 1961 827. 1109.8 1445.2 728.9 741.3 1465 1005.
xen-1cpu Linux 2.6.18- 1768 2002 830. 1108.5 1447.9 731.0 744.8 1470 1009.
xen-1cpu Linux 2.6.18- 1764 1937 813. 1107.7 1453.0 735.3 746.8 1473 1008.
xen-1cpu Linux 2.6.18- 1764 1959 808. 1108.2 1457.6 734.2 748.8 1476 1017.
nativ-1cp Linux 2.6.18- 2025 2274 777. 1166.8 1447.4 820.8 872.0 1471 1088.
nativ-1cp Linux 2.6.18- 1988 2392 717. 1162.2 1447.0 820.0 871.3 1471 1089.
nativ-1cp Linux 2.6.18- 2054 2255 784. 1162.4 1447.6 820.9 871.9 1470 1089.
nativ-1cp Linux 2.6.18- 2044 2242 743. 1159.4 1446.4 820.0 871.2 1471 1090.
nativ-1cp Linux 2.6.18- 2050 2233 727. 1159.8 1446.9 820.4 871.4 1471 1090.
nativ-1cp Linux 2.6.18- 2022 2271 776. 1160.9 1448.8 819.6 871.3 1471 1089.
nativ-1cp Linux 2.6.18- 1970 2258 750. 1156.8 1446.7 821.0 871.9 1472 1089.
nativ-1cp Linux 2.6.18- 2013 2236 773. 1161.4 1452.6 822.2 874.2 1476 1094.
nativ-1cp Linux 2.6.18- 2018 2273 763. 1158.8 1447.0 820.1 873.1 1472 1090.
nativ-1cp Linux 2.6.18- 2026 2230 773. 1157.1 1448.2 818.6 871.2 1472 1090.
Memory latencies in nanoseconds - smaller is better
(WARNING - may not be correct, check graphs)
------------------------------------------------------------------------------
Host OS Mhz L1 $ L2 $ Main mem Rand mem Guesses
--------- ------------- --- ---- ---- -------- -------- -------
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4630 138.4 238.0
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4590 153.3 246.5
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4590 152.8 247.9
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4610 152.5 251.5
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4600 152.2 255.2
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4590 151.8 255.6
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4650 151.6 252.0
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4600 151.3 258.4
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4600 118.9 256.0
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4630 150.7 253.6
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4620 164.7 237.4
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4640 164.7 244.2
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4630 164.7 239.9
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4600 164.6 243.0
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4590 164.8 241.8
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4640 164.7 245.2
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4640 164.7 249.4
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4630 163.9 249.9
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4590 164.6 251.5
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4650 164.6 250.8
make[1]: Leaving directory `/benchmark/lmbench3/results'
[-- Attachment #3: Type: text/plain, Size: 137 bytes --]
_______________________________________________
Xen-users mailing list
Xen-users@lists.xensource.com
http://lists.xensource.com/xen-users
^ permalink raw reply [flat|nested] 16+ messages in thread
* Benchmark results
@ 2008-08-03 16:51 bbmailing
2008-08-03 17:07 ` Keir Fraser
0 siblings, 1 reply; 16+ messages in thread
From: bbmailing @ 2008-08-03 16:51 UTC (permalink / raw)
To: xen-devel
[-- Attachment #1: Type: text/plain, Size: 1816 bytes --]
Hey everyone,
I hope you guys can help me explaining a pretty strange result that came up when benchmarking a native Red Hat installation against a Xen DomU (installed in a file container). I used lmbench3 on a dualcore AMD-machine. To make the results comparable, I "unplugged" all CPU cores but one in the native installation (setting the "online"-entry to 0). The DomU had only one vCPU assigned. The installed packages on both machines were identical as well. After doing the initial configuration of lmbench3 I ran a script doing 10 times a "make rerun".
All results seem to be OK somehow. But I cannot explain why Xen is faster than the native installation when creating and deleting files? Its 12ms (DomU) against 24ms (native) for creating a 0K file and 50ms (DomU) against 60ms (native) for creating a 10K file. Please find all the results attached.
Is there any way this is possible? Maybe the communication methods used by the shared device drivers make the virtual machine think the operation has been finished successfully faster than the native way? Is there is anyone out there who can help me understanding this result?
Furthermore: Is there a special recommendation for benchmarking Xen against native installations? I liked lmbench so far since it puts the focus on basic operations... what are your favourites? Is lmbench from a design perspective the wrong tool to use?
I have to find some reliable results so I am really thankful for every single hint here... please feel free to email me directly as well!
Thank you so much in advance!
Best regards,
Bjoern
_______________________________________________________________________
Jetzt neu! Schützen Sie Ihren PC mit McAfee und WEB.DE. 30 Tage
kostenlos testen. http://www.pc-sicherheit.web.de/startseite/?mc=022220
[-- Attachment #2: readable.txt --]
[-- Type: text/plain, Size: 17504 bytes --]
cd results && make summary percent 2>/dev/null | more
make[1]: Entering directory `/benchmark/lmbench3/results'
L M B E N C H 3 . 0 S U M M A R Y
------------------------------------
(Alpha software, do not distribute)
Basic system parameters
------------------------------------------------------------------------------
Host OS Description Mhz tlb cache mem scal
pages line par load
bytes
--------- ------------- ----------------------- ---- ----- ----- ------ ----
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.2800 1
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.0300 1
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.0200 1
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.0200 1
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.1300 1
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.0200 1
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.1400 1
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.0100 1
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.1500 1
xen-1cpu Linux 2.6.18- 1cpu 2800 32 128 1.0100 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0100 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0100 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0200 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0100 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0100 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0100 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0100 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0100 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0100 1
nativ-1cp Linux 2.6.18- nativ-1cpu 2800 32 128 1.0100 1
Processor, Processes - times in microseconds - smaller is better
------------------------------------------------------------------------------
Host OS Mhz null null open slct sig sig fork exec sh
call I/O stat clos TCP inst hndl proc proc proc
--------- ------------- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ----
xen-1cpu Linux 2.6.18- 2800 0.87 1.29 2.69 5.45 5.20 1.08 2.78 418. 1043 3264
xen-1cpu Linux 2.6.18- 2800 0.85 1.29 2.64 5.53 5.25 1.03 2.71 467. 1092 3159
xen-1cpu Linux 2.6.18- 2800 0.85 1.27 2.68 5.58 5.18 1.04 2.74 465. 1103 3121
xen-1cpu Linux 2.6.18- 2800 0.82 1.27 2.65 5.52 5.22 1.03 2.69 461. 1102 3146
xen-1cpu Linux 2.6.18- 2800 0.84 1.28 2.64 5.69 5.24 1.05 2.79 456. 1090 3119
xen-1cpu Linux 2.6.18- 2800 0.84 1.28 2.64 5.53 5.20 1.02 2.72 455. 1094 3125
xen-1cpu Linux 2.6.18- 2800 0.85 1.29 2.62 5.50 5.22 1.01 2.80 464. 1089 3147
xen-1cpu Linux 2.6.18- 2800 0.84 1.31 2.65 5.50 5.22 1.00 2.72 463. 1095 3130
xen-1cpu Linux 2.6.18- 2800 0.84 1.29 2.69 5.51 5.20 1.04 2.74 457. 1093 3138
xen-1cpu Linux 2.6.18- 2800 0.84 1.29 2.66 5.35 5.21 1.03 2.78 469. 1095 3133
nativ-1cp Linux 2.6.18- 2800 0.19 0.44 2.02 3.45 6.92 0.28 1.25 125. 407. 1905
nativ-1cp Linux 2.6.18- 2800 0.19 0.51 2.19 3.57 10.7 0.28 1.25 126. 404. 1903
nativ-1cp Linux 2.6.18- 2800 0.19 0.49 1.84 3.12 10.6 0.28 1.40 129. 398. 1903
nativ-1cp Linux 2.6.18- 2800 0.19 0.46 2.03 3.31 4.80 0.28 1.32 125. 403. 1910
nativ-1cp Linux 2.6.18- 2800 0.19 0.50 2.04 3.35 9.60 0.28 1.18 123. 403. 1905
nativ-1cp Linux 2.6.18- 2800 0.19 0.47 1.96 3.29 4.57 0.28 1.25 126. 399. 1894
nativ-1cp Linux 2.6.18- 2800 0.19 0.45 1.95 3.22 7.39 0.28 1.26 123. 398. 1925
nativ-1cp Linux 2.6.18- 2800 0.19 0.45 2.16 3.28 4.56 0.28 1.37 129. 400. 1919
nativ-1cp Linux 2.6.18- 2800 0.19 0.43 1.92 3.10 9.63 0.28 1.34 128. 413. 1954
nativ-1cp Linux 2.6.18- 2800 0.24 0.47 1.84 3.33 4.49 0.33 1.49 125. 410. 1931
Basic integer operations - times in nanoseconds - smaller is better
-------------------------------------------------------------------
Host OS intgr intgr intgr intgr intgr
bit add mul div mod
--------- ------------- ------ ------ ------ ------ ------
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
xen-1cpu Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.1
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.1
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.1
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.1
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
nativ-1cp Linux 2.6.18- 0.3600 0.0100 0.1600 15.4 15.0
Basic float operations - times in nanoseconds - smaller is better
-----------------------------------------------------------------
Host OS float float float float
add mul div bogo
--------- ------------- ------ ------ ------ ------
xen-1cpu Linux 2.6.18- 1.4400 1.4500 6.3500 5.5900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 6.3500 5.5900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 6.3300 5.5900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 6.3200 5.5900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 6.3500 5.5900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 6.3200 5.6200
xen-1cpu Linux 2.6.18- 1.4300 1.4500 6.3500 5.5900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 6.3300 5.5900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 6.3500 5.5900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 6.3500 5.5900
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3500 5.6500
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3400 5.6700
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3500 5.6200
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3500 5.6400
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3500 5.6200
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3700 5.6500
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3500 5.6300
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3500 5.6600
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3500 5.6300
nativ-1cp Linux 2.6.18- 1.4300 1.4500 6.3600 5.6600
Basic double operations - times in nanoseconds - smaller is better
------------------------------------------------------------------
Host OS double double double double
add mul div bogo
--------- ------------- ------ ------ ------ ------
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7600 7.2900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7800 7.2900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7700 7.2800
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7700 7.2800
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7600 7.2900
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7800 7.2800
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7800 7.2800
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7600 7.2800
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7800 7.2800
xen-1cpu Linux 2.6.18- 1.4300 1.4500 7.7900 7.3100
nativ-1cp Linux 2.6.18- 1.4400 1.4500 7.7800 7.3000
nativ-1cp Linux 2.6.18- 1.4300 1.4500 7.7800 7.3000
nativ-1cp Linux 2.6.18- 1.4300 1.4500 7.7800 7.3000
nativ-1cp Linux 2.6.18- 1.4300 1.4500 7.7800 7.3000
nativ-1cp Linux 2.6.18- 1.4300 1.4500 7.7800 7.3000
nativ-1cp Linux 2.6.18- 1.4300 1.4500 7.7900 7.3000
nativ-1cp Linux 2.6.18- 1.4300 1.4500 7.8200 7.3000
nativ-1cp Linux 2.6.18- 1.4300 1.4500 7.7800 7.3000
nativ-1cp Linux 2.6.18- 1.4300 1.4500 7.7800 7.3000
nativ-1cp Linux 2.6.18- 1.4300 1.4500 7.7900 7.3200
Context switching - times in microseconds - smaller is better
-------------------------------------------------------------------------
Host OS 2p/0K 2p/16K 2p/64K 8p/16K 8p/64K 16p/16K 16p/64K
ctxsw ctxsw ctxsw ctxsw ctxsw ctxsw ctxsw
--------- ------------- ------ ------ ------ ------ ------ ------- -------
xen-1cpu Linux 2.6.18- 2.4900 2.7300 5.2100 4.2200 7.2100 4.63000 38.6
xen-1cpu Linux 2.6.18- 2.5800 2.7300 5.2600 4.1400 8.0800 4.90000 47.2
xen-1cpu Linux 2.6.18- 2.5800 2.6400 5.2400 4.1100 7.5000 4.09000 46.4
xen-1cpu Linux 2.6.18- 2.7100 2.7800 5.2500 4.2000 8.0400 5.10000 43.5
xen-1cpu Linux 2.6.18- 2.5400 2.7500 5.3900 4.1100 7.6700 4.84000 44.0
xen-1cpu Linux 2.6.18- 2.5200 2.7400 5.1500 4.2100 6.7300 4.60000 40.1
xen-1cpu Linux 2.6.18- 2.6100 2.6600 5.3000 4.1800 7.5800 4.68000 41.5
xen-1cpu Linux 2.6.18- 2.5900 2.7400 5.2100 4.2000 6.8300 4.65000 43.0
xen-1cpu Linux 2.6.18- 2.5100 2.6300 5.2100 4.0400 6.6800 4.74000 43.8
xen-1cpu Linux 2.6.18- 2.5700 2.8100 5.1000 4.2200 8.5200 4.81000 43.2
nativ-1cp Linux 2.6.18- 0.7400 0.8200 3.2200 1.7600 4.1600 1.91000 37.3
nativ-1cp Linux 2.6.18- 0.6000 0.8600 3.0400 1.8100 4.8200 1.78000 38.4
nativ-1cp Linux 2.6.18- 0.6500 0.7600 3.2500 1.8200 3.8800 1.82000 40.3
nativ-1cp Linux 2.6.18- 0.6200 0.8000 3.3900 1.7700 7.1800 2.28000 38.8
nativ-1cp Linux 2.6.18- 0.7800 0.8500 3.3700 1.8600 3.7200 1.97000 36.4
nativ-1cp Linux 2.6.18- 0.6400 0.8100 3.3600 1.8100 4.3400 2.46000 36.0
nativ-1cp Linux 2.6.18- 0.5600 0.7500 3.0700 1.7400 4.1200 2.23000 33.0
nativ-1cp Linux 2.6.18- 0.6500 0.9100 3.3100 1.8700 4.3100 1.90000 37.5
nativ-1cp Linux 2.6.18- 0.6100 0.7900 3.4100 1.7200 4.5600 1.86000 33.8
nativ-1cp Linux 2.6.18- 0.5800 0.8300 3.1500 1.8200 4.1600 2.28000 37.9
*Local* Communication latencies in microseconds - smaller is better
---------------------------------------------------------------------
Host OS 2p/0K Pipe AF UDP RPC/ TCP RPC/ TCP
ctxsw UNIX UDP TCP conn
--------- ------------- ----- ----- ---- ----- ----- ----- ----- ----
xen-1cpu Linux 2.6.18- 2.490 12.1 19.5 17.6 23.7 20.8 29.2 54.
xen-1cpu Linux 2.6.18- 2.580 12.0 19.8 17.4 23.7 20.9 28.8 55.
xen-1cpu Linux 2.6.18- 2.580 12.0 19.2 17.4 23.6 21.0 28.5 55.
xen-1cpu Linux 2.6.18- 2.710 12.1 19.3 17.4 23.6 21.1 28.3 55.
xen-1cpu Linux 2.6.18- 2.540 12.9 19.3 17.4 23.7 21.2 28.7 55.
xen-1cpu Linux 2.6.18- 2.520 12.2 19.4 17.4 23.6 21.1 28.4 55.
xen-1cpu Linux 2.6.18- 2.610 12.9 19.4 17.4 23.7 21.2 28.8 55.
xen-1cpu Linux 2.6.18- 2.590 13.0 19.5 17.6 23.7 21.1 28.7 55.
xen-1cpu Linux 2.6.18- 2.510 12.0 19.3 17.2 23.8 21.3 28.6 55.
xen-1cpu Linux 2.6.18- 2.570 13.0 19.2 17.5 23.9 21.1 28.6 55.
nativ-1cp Linux 2.6.18- 0.740 4.631 7.59 9.379 12.8 12.1 16.7 126.
nativ-1cp Linux 2.6.18- 0.600 4.756 8.19 9.929 12.9 12.4 17.2 131.
nativ-1cp Linux 2.6.18- 0.650 4.589 7.62 9.214 12.7 12.2 16.7 123.
nativ-1cp Linux 2.6.18- 0.620 4.587 7.97 9.319 13.1 12.1 17.7 126.
nativ-1cp Linux 2.6.18- 0.780 4.582 8.30 10.0 13.1 12.6 16.8 128.
nativ-1cp Linux 2.6.18- 0.640 4.628 7.88 9.579 13.9 12.3 17.7 130.
nativ-1cp Linux 2.6.18- 0.560 4.575 7.80 9.656 13.2 12.5 17.5 125.
nativ-1cp Linux 2.6.18- 0.650 4.986 8.12 9.576 12.9 12.3 17.3 127.
nativ-1cp Linux 2.6.18- 0.610 4.635 7.69 10.0 13.1 12.2 16.7 125.
nativ-1cp Linux 2.6.18- 0.580 4.782 7.80 9.993 13.8 12.1 17.7 126.
File & VM system latencies in microseconds - smaller is better
-------------------------------------------------------------------------------
Host OS 0K File 10K File Mmap Prot Page 100fd
Create Delete Create Delete Latency Fault Fault selct
--------- ------------- ------ ------ ------ ------ ------- ----- ------- -----
xen-1cpu Linux 2.6.18- 11.9 9.2752 51.7 18.5 14.3K 0.848 3.97750 2.628
xen-1cpu Linux 2.6.18- 12.3 9.3171 53.8 19.4 15.4K 0.864 4.06550 2.619
xen-1cpu Linux 2.6.18- 12.3 9.6721 53.7 19.1 15.5K 0.910 4.16050 2.596
xen-1cpu Linux 2.6.18- 12.1 9.5280 41.9 19.1 15.7K 0.944 4.19680 2.588
xen-1cpu Linux 2.6.18- 12.1 9.4666 46.8 19.8 15.8K 0.892 4.18000 2.638
xen-1cpu Linux 2.6.18- 12.1 9.4604 53.5 19.7 16.5K 0.918 4.22680 2.598
xen-1cpu Linux 2.6.18- 12.3 9.7807 48.2 19.6 16.0K 0.905 4.25190 2.625
xen-1cpu Linux 2.6.18- 12.2 9.7652 52.0 20.4 16.0K 0.896 4.23290 2.695
xen-1cpu Linux 2.6.18- 12.3 9.6259 53.6 20.1 16.0K 0.867 4.26970 2.629
xen-1cpu Linux 2.6.18- 12.2 9.6403 52.7 19.6 16.1K 0.861 4.27980 2.613
nativ-1cp Linux 2.6.18- 24.8 14.5 59.5 26.4 5090.0 0.554 1.52850 4.263
nativ-1cp Linux 2.6.18- 26.3 16.1 58.7 27.3 5118.0 0.585 1.71360 4.251
nativ-1cp Linux 2.6.18- 24.9 15.4 60.5 26.3 4892.0 0.612 1.52190 1.942
nativ-1cp Linux 2.6.18- 24.4 16.1 58.7 27.2 5277.0 0.373 1.66030 1.953
nativ-1cp Linux 2.6.18- 24.3 14.7 59.5 26.5 5130.0 0.627 1.54020 1.924
nativ-1cp Linux 2.6.18- 23.9 16.1 57.8 26.1 5080.0 0.571 1.55070 1.980
nativ-1cp Linux 2.6.18- 25.6 15.5 68.7 26.8 5214.0 0.502 1.75650 1.941
nativ-1cp Linux 2.6.18- 24.7 14.4 58.3 25.4 5014.0 0.552 1.73150 1.965
nativ-1cp Linux 2.6.18- 25.6 15.2 60.4 26.4 5080.0 0.321 1.67840 1.973
nativ-1cp Linux 2.6.18- 25.8 15.2 61.6 27.0 5221.0 0.694 1.71090 1.967
*Local* Communication bandwidths in MB/s - bigger is better
-----------------------------------------------------------------------------
Host OS Pipe AF TCP File Mmap Bcopy Bcopy Mem Mem
UNIX reread reread (libc) (hand) read write
--------- ------------- ---- ---- ---- ------ ------ ------ ------ ---- -----
xen-1cpu Linux 2.6.18- 1764 2004 812. 1285.7 1889.5 834.1 819.5 1644 1121.
xen-1cpu Linux 2.6.18- 1762 1976 797. 1120.6 1405.6 705.4 724.1 1436 966.3
xen-1cpu Linux 2.6.18- 1772 1997 852. 1119.6 1420.5 718.9 727.7 1446 987.7
xen-1cpu Linux 2.6.18- 1758 2004 828. 1115.8 1429.3 720.6 730.2 1448 991.7
xen-1cpu Linux 2.6.18- 1749 1946 827. 1109.9 1434.8 721.6 733.2 1457 996.8
xen-1cpu Linux 2.6.18- 1777 1987 818. 1113.4 1440.1 725.4 742.8 1462 1001.
xen-1cpu Linux 2.6.18- 1761 1961 827. 1109.8 1445.2 728.9 741.3 1465 1005.
xen-1cpu Linux 2.6.18- 1768 2002 830. 1108.5 1447.9 731.0 744.8 1470 1009.
xen-1cpu Linux 2.6.18- 1764 1937 813. 1107.7 1453.0 735.3 746.8 1473 1008.
xen-1cpu Linux 2.6.18- 1764 1959 808. 1108.2 1457.6 734.2 748.8 1476 1017.
nativ-1cp Linux 2.6.18- 2025 2274 777. 1166.8 1447.4 820.8 872.0 1471 1088.
nativ-1cp Linux 2.6.18- 1988 2392 717. 1162.2 1447.0 820.0 871.3 1471 1089.
nativ-1cp Linux 2.6.18- 2054 2255 784. 1162.4 1447.6 820.9 871.9 1470 1089.
nativ-1cp Linux 2.6.18- 2044 2242 743. 1159.4 1446.4 820.0 871.2 1471 1090.
nativ-1cp Linux 2.6.18- 2050 2233 727. 1159.8 1446.9 820.4 871.4 1471 1090.
nativ-1cp Linux 2.6.18- 2022 2271 776. 1160.9 1448.8 819.6 871.3 1471 1089.
nativ-1cp Linux 2.6.18- 1970 2258 750. 1156.8 1446.7 821.0 871.9 1472 1089.
nativ-1cp Linux 2.6.18- 2013 2236 773. 1161.4 1452.6 822.2 874.2 1476 1094.
nativ-1cp Linux 2.6.18- 2018 2273 763. 1158.8 1447.0 820.1 873.1 1472 1090.
nativ-1cp Linux 2.6.18- 2026 2230 773. 1157.1 1448.2 818.6 871.2 1472 1090.
Memory latencies in nanoseconds - smaller is better
(WARNING - may not be correct, check graphs)
------------------------------------------------------------------------------
Host OS Mhz L1 $ L2 $ Main mem Rand mem Guesses
--------- ------------- --- ---- ---- -------- -------- -------
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4630 138.4 238.0
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4590 153.3 246.5
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4590 152.8 247.9
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4610 152.5 251.5
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4600 152.2 255.2
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4590 151.8 255.6
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4650 151.6 252.0
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4600 151.3 258.4
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4600 118.9 256.0
xen-1cpu Linux 2.6.18- 2800 1.0720 4.4630 150.7 253.6
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4620 164.7 237.4
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4640 164.7 244.2
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4630 164.7 239.9
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4600 164.6 243.0
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4590 164.8 241.8
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4640 164.7 245.2
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4640 164.7 249.4
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4630 163.9 249.9
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4590 164.6 251.5
nativ-1cp Linux 2.6.18- 2800 1.0740 4.4650 164.6 250.8
make[1]: Leaving directory `/benchmark/lmbench3/results'
[-- Attachment #3: Type: text/plain, Size: 138 bytes --]
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: Benchmark results
2008-08-03 16:51 bbmailing
@ 2008-08-03 17:07 ` Keir Fraser
0 siblings, 0 replies; 16+ messages in thread
From: Keir Fraser @ 2008-08-03 17:07 UTC (permalink / raw)
To: bbmailing, xen-devel
If the virtual disc is a loopback-mounted file then the domU is probably
benefitting from dom0's buffer cache (i.e., domU is happening to exploit
dom0's memory to cache its own data).
-- Keir
On 3/8/08 17:51, "bbmailing@web.de" <bbmailing@web.de> wrote:
> Hey everyone,
>
> I hope you guys can help me explaining a pretty strange result that came up
> when benchmarking a native Red Hat installation against a Xen DomU (installed
> in a file container). I used lmbench3 on a dualcore AMD-machine. To make the
> results comparable, I "unplugged" all CPU cores but one in the native
> installation (setting the "online"-entry to 0). The DomU had only one vCPU
> assigned. The installed packages on both machines were identical as well.
> After doing the initial configuration of lmbench3 I ran a script doing 10
> times a "make rerun".
>
> All results seem to be OK somehow. But I cannot explain why Xen is faster than
> the native installation when creating and deleting files? Its 12ms (DomU)
> against 24ms (native) for creating a 0K file and 50ms (DomU) against 60ms
> (native) for creating a 10K file. Please find all the results attached.
>
> Is there any way this is possible? Maybe the communication methods used by the
> shared device drivers make the virtual machine think the operation has been
> finished successfully faster than the native way? Is there is anyone out there
> who can help me understanding this result?
>
> Furthermore: Is there a special recommendation for benchmarking Xen against
> native installations? I liked lmbench so far since it puts the focus on basic
> operations... what are your favourites? Is lmbench from a design perspective
> the wrong tool to use?
>
> I have to find some reliable results so I am really thankful for every single
> hint here... please feel free to email me directly as well!
>
> Thank you so much in advance!
>
> Best regards,
> Bjoern
> _______________________________________________________________________
> Jetzt neu! Schützen Sie Ihren PC mit McAfee und WEB.DE. 30 Tage
> kostenlos testen. http://www.pc-sicherheit.web.de/startseite/?mc=022220
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel
^ permalink raw reply [flat|nested] 16+ messages in thread
* benchmark results
@ 2009-12-24 10:31 ` Christian Kujau
0 siblings, 0 replies; 16+ messages in thread
From: Christian Kujau @ 2009-12-24 10:31 UTC (permalink / raw)
To: xfs, reiserfs-devel, linux-ext4, linux-btrfs, jfs-discussion,
ext-users, linu
I've had the chance to use a testsystem here and couldn't resist running a
few benchmark programs on them: bonnie++, tiobench, dbench and a few
generic ones (cp/rm/tar/etc...) on ext{234}, btrfs, jfs, ufs, xfs, zfs.
All with standard mkfs/mount options and +noatime for all of them.
Here are the results, no graphs - sorry:
http://nerdbynature.de/benchmarks/v40z/2009-12-22/
Reiserfs is locking up during dbench, so I removed it from the
config, here are some earlier results:
http://nerdbynature.de/benchmarks/v40z/2009-12-21/bonnie.html
Bonnie++ couldn't complete on nilfs2, only the generic tests
and tiobench were run. As nilfs2, ufs, zfs aren't supporting xattr, dbench
could not be run on these filesystems.
Short summary, AFAICT:
- btrfs, ext4 are the overall winners
- xfs to, but creating/deleting many files was *very* slow
- if you need only fast but no cool features or journaling, ext2
is still a good choice :)
Thanks,
Christian.
--
BOFH excuse #84:
Someone is standing on the ethernet cable, causing a kink in the cable
^ permalink raw reply [flat|nested] 16+ messages in thread
* benchmark results
@ 2009-12-24 10:31 ` Christian Kujau
0 siblings, 0 replies; 16+ messages in thread
From: Christian Kujau @ 2009-12-24 10:31 UTC (permalink / raw)
To: xfs, reiserfs-devel, linux-ext4, linux-btrfs, jfs-discussion,
ext-users, linux-nilfs
I've had the chance to use a testsystem here and couldn't resist running a
few benchmark programs on them: bonnie++, tiobench, dbench and a few
generic ones (cp/rm/tar/etc...) on ext{234}, btrfs, jfs, ufs, xfs, zfs.
All with standard mkfs/mount options and +noatime for all of them.
Here are the results, no graphs - sorry:
http://nerdbynature.de/benchmarks/v40z/2009-12-22/
Reiserfs is locking up during dbench, so I removed it from the
config, here are some earlier results:
http://nerdbynature.de/benchmarks/v40z/2009-12-21/bonnie.html
Bonnie++ couldn't complete on nilfs2, only the generic tests
and tiobench were run. As nilfs2, ufs, zfs aren't supporting xattr, dbench
could not be run on these filesystems.
Short summary, AFAICT:
- btrfs, ext4 are the overall winners
- xfs to, but creating/deleting many files was *very* slow
- if you need only fast but no cool features or journaling, ext2
is still a good choice :)
Thanks,
Christian.
--
BOFH excuse #84:
Someone is standing on the ethernet cable, causing a kink in the cable
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: benchmark results
2009-12-24 10:31 ` Christian Kujau
@ 2009-12-24 12:06 ` Ryusuke Konishi
-1 siblings, 0 replies; 16+ messages in thread
From: Ryusuke Konishi @ 2009-12-24 12:06 UTC (permalink / raw)
To: lists
Cc: xfs, reiserfs-devel, linux-ext4, linux-btrfs, jfs-discussion,
ext3-users, linux-nilfs
Hi,
On Thu, 24 Dec 2009 02:31:10 -0800 (PST), Christian Kujau wrote:
> I've had the chance to use a testsystem here and couldn't resist running a
> few benchmark programs on them: bonnie++, tiobench, dbench and a few
> generic ones (cp/rm/tar/etc...) on ext{234}, btrfs, jfs, ufs, xfs, zfs.
>
> All with standard mkfs/mount options and +noatime for all of them.
>
> Here are the results, no graphs - sorry:
> http://nerdbynature.de/benchmarks/v40z/2009-12-22/
>
> Reiserfs is locking up during dbench, so I removed it from the
> config, here are some earlier results:
>
> http://nerdbynature.de/benchmarks/v40z/2009-12-21/bonnie.html
>
> Bonnie++ couldn't complete on nilfs2, only the generic tests
> and tiobench were run.
I looked at the log but couldn't identify the error.
Is that a disk full?
Thanks,
Ryusuke Konishi
> As nilfs2, ufs, zfs aren't supporting xattr, dbench
> could not be run on these filesystems.
>
> Short summary, AFAICT:
> - btrfs, ext4 are the overall winners
> - xfs to, but creating/deleting many files was *very* slow
> - if you need only fast but no cool features or journaling, ext2
> is still a good choice :)
>
> Thanks,
> Christian.
> --
> BOFH excuse #84:
>
> Someone is standing on the ethernet cable, causing a kink in the cable
> --
> To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: benchmark results
@ 2009-12-24 12:06 ` Ryusuke Konishi
0 siblings, 0 replies; 16+ messages in thread
From: Ryusuke Konishi @ 2009-12-24 12:06 UTC (permalink / raw)
To: lists
Cc: jfs-discussion, linux-nilfs, reiserfs-devel, xfs, ext3-users,
linux-ext4, linux-btrfs
Hi,
On Thu, 24 Dec 2009 02:31:10 -0800 (PST), Christian Kujau wrote:
> I've had the chance to use a testsystem here and couldn't resist running a
> few benchmark programs on them: bonnie++, tiobench, dbench and a few
> generic ones (cp/rm/tar/etc...) on ext{234}, btrfs, jfs, ufs, xfs, zfs.
>
> All with standard mkfs/mount options and +noatime for all of them.
>
> Here are the results, no graphs - sorry:
> http://nerdbynature.de/benchmarks/v40z/2009-12-22/
>
> Reiserfs is locking up during dbench, so I removed it from the
> config, here are some earlier results:
>
> http://nerdbynature.de/benchmarks/v40z/2009-12-21/bonnie.html
>
> Bonnie++ couldn't complete on nilfs2, only the generic tests
> and tiobench were run.
I looked at the log but couldn't identify the error.
Is that a disk full?
Thanks,
Ryusuke Konishi
> As nilfs2, ufs, zfs aren't supporting xattr, dbench
> could not be run on these filesystems.
>
> Short summary, AFAICT:
> - btrfs, ext4 are the overall winners
> - xfs to, but creating/deleting many files was *very* slow
> - if you need only fast but no cool features or journaling, ext2
> is still a good choice :)
>
> Thanks,
> Christian.
> --
> BOFH excuse #84:
>
> Someone is standing on the ethernet cable, causing a kink in the cable
> --
> To unsubscribe from this list: send the line "unsubscribe linux-nilfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: benchmark results
2009-12-24 10:31 ` Christian Kujau
(?)
@ 2009-12-24 12:59 ` Teran McKinney
-1 siblings, 0 replies; 16+ messages in thread
From: Teran McKinney @ 2009-12-24 12:59 UTC (permalink / raw)
To: Christian Kujau
Cc: xfs, reiserfs-devel, linux-ext4, linux-btrfs, jfs-discussion,
ext-users, linux-nilfs
Which I/O scheduler are you using? Pretty sure that ReiserFS is a
little less deadlocky with CFQ or another over deadline, but that
deadline usually gives the best results for me (especially for JFS).
Thanks,
Teran
On Thu, Dec 24, 2009 at 10:31, Christian Kujau <lists@nerdbynature.de> =
wrote:
> I've had the chance to use a testsystem here and couldn't resist runn=
ing a
> few benchmark programs on them: bonnie++, tiobench, dbench and a few
> generic ones (cp/rm/tar/etc...) on ext{234}, btrfs, jfs, ufs, xfs, zf=
s.
>
> All with standard mkfs/mount options and +noatime for all of them.
>
> Here are the results, no graphs - sorry:
> =C2=A0 http://nerdbynature.de/benchmarks/v40z/2009-12-22/
>
> Reiserfs is locking up during dbench, so I removed it from the
> config, here are some earlier results:
>
> =C2=A0 http://nerdbynature.de/benchmarks/v40z/2009-12-21/bonnie.html
>
> Bonnie++ couldn't complete on nilfs2, only the generic tests
> and tiobench were run. As nilfs2, ufs, zfs aren't supporting xattr, d=
bench
> could not be run on these filesystems.
>
> Short summary, AFAICT:
> =C2=A0 =C2=A0- btrfs, ext4 are the overall winners
> =C2=A0 =C2=A0- xfs to, but creating/deleting many files was *very* sl=
ow
> =C2=A0 =C2=A0- if you need only fast but no cool features or journali=
ng, ext2
> =C2=A0 =C2=A0 =C2=A0is still a good choice :)
>
> Thanks,
> Christian.
> --
> BOFH excuse #84:
>
> Someone is standing on the ethernet cable, causing a kink in the cabl=
e
> --
> To unsubscribe from this list: send the line "unsubscribe reiserfs-de=
vel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at =C2=A0http://vger.kernel.org/majordomo-info.ht=
ml
>
--
To unsubscribe from this list: send the line "unsubscribe reiserfs-deve=
l" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: benchmark results
@ 2009-12-24 12:59 ` Teran McKinney
0 siblings, 0 replies; 16+ messages in thread
From: Teran McKinney @ 2009-12-24 12:59 UTC (permalink / raw)
To: Christian Kujau
Cc: xfs, reiserfs-devel, linux-ext4, linux-btrfs, jfs-discussion,
ext-users, linux-nilfs
Which I/O scheduler are you using? Pretty sure that ReiserFS is a
little less deadlocky with CFQ or another over deadline, but that
deadline usually gives the best results for me (especially for JFS).
Thanks,
Teran
On Thu, Dec 24, 2009 at 10:31, Christian Kujau <lists@nerdbynature.de> wrote:
> I've had the chance to use a testsystem here and couldn't resist running a
> few benchmark programs on them: bonnie++, tiobench, dbench and a few
> generic ones (cp/rm/tar/etc...) on ext{234}, btrfs, jfs, ufs, xfs, zfs.
>
> All with standard mkfs/mount options and +noatime for all of them.
>
> Here are the results, no graphs - sorry:
> http://nerdbynature.de/benchmarks/v40z/2009-12-22/
>
> Reiserfs is locking up during dbench, so I removed it from the
> config, here are some earlier results:
>
> http://nerdbynature.de/benchmarks/v40z/2009-12-21/bonnie.html
>
> Bonnie++ couldn't complete on nilfs2, only the generic tests
> and tiobench were run. As nilfs2, ufs, zfs aren't supporting xattr, dbench
> could not be run on these filesystems.
>
> Short summary, AFAICT:
> - btrfs, ext4 are the overall winners
> - xfs to, but creating/deleting many files was *very* slow
> - if you need only fast but no cool features or journaling, ext2
> is still a good choice :)
>
> Thanks,
> Christian.
> --
> BOFH excuse #84:
>
> Someone is standing on the ethernet cable, causing a kink in the cable
> --
> To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
--
To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: benchmark results
@ 2009-12-24 12:59 ` Teran McKinney
0 siblings, 0 replies; 16+ messages in thread
From: Teran McKinney @ 2009-12-24 12:59 UTC (permalink / raw)
To: Christian Kujau
Cc: jfs-discussion, linux-nilfs, reiserfs-devel, xfs, ext-users,
linux-ext4, linux-btrfs
Which I/O scheduler are you using? Pretty sure that ReiserFS is a
little less deadlocky with CFQ or another over deadline, but that
deadline usually gives the best results for me (especially for JFS).
Thanks,
Teran
On Thu, Dec 24, 2009 at 10:31, Christian Kujau <lists@nerdbynature.de> wrote:
> I've had the chance to use a testsystem here and couldn't resist running a
> few benchmark programs on them: bonnie++, tiobench, dbench and a few
> generic ones (cp/rm/tar/etc...) on ext{234}, btrfs, jfs, ufs, xfs, zfs.
>
> All with standard mkfs/mount options and +noatime for all of them.
>
> Here are the results, no graphs - sorry:
> http://nerdbynature.de/benchmarks/v40z/2009-12-22/
>
> Reiserfs is locking up during dbench, so I removed it from the
> config, here are some earlier results:
>
> http://nerdbynature.de/benchmarks/v40z/2009-12-21/bonnie.html
>
> Bonnie++ couldn't complete on nilfs2, only the generic tests
> and tiobench were run. As nilfs2, ufs, zfs aren't supporting xattr, dbench
> could not be run on these filesystems.
>
> Short summary, AFAICT:
> - btrfs, ext4 are the overall winners
> - xfs to, but creating/deleting many files was *very* slow
> - if you need only fast but no cool features or journaling, ext2
> is still a good choice :)
>
> Thanks,
> Christian.
> --
> BOFH excuse #84:
>
> Someone is standing on the ethernet cable, causing a kink in the cable
> --
> To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: benchmark results
2009-12-24 10:31 ` Christian Kujau
` (2 preceding siblings ...)
(?)
@ 2009-12-24 13:05 ` Peter Grandi
-1 siblings, 0 replies; 16+ messages in thread
From: Peter Grandi @ 2009-12-24 13:05 UTC (permalink / raw)
To: xfs, reiserfs-devel, linux-ext4, linux-btrfs, jfs-discussion,
ext-users, linu
> I've had the chance to use a testsystem here and couldn't
> resist
Unfortunately there seems to be an overproduction of rather
meaningless file system "benchmarks"...
> running a few benchmark programs on them: bonnie++, tiobench,
> dbench and a few generic ones (cp/rm/tar/etc...) on ext{234},
> btrfs, jfs, ufs, xfs, zfs. All with standard mkfs/mount options
> and +noatime for all of them.
> Here are the results, no graphs - sorry: [ ... ]
After having a glance, I suspect that your tests could be
enormously improved, and doing so would reduce the pointlessness of
the results.
A couple of hints:
* In the "generic" test the 'tar' test bandwidth is exactly the
same ("276.68 MB/s") for nearly all filesystems.
* There are read transfer rates higher than the one reported by
'hdparm' which is "66.23 MB/sec" (comically enough *all* the
read transfer rates your "benchmarks" report are higher).
BTW the use of Bonnie++ is also usually a symptom of a poor
misunderstanding of file system benchmarking.
On the plus side, test setup context is provided in the "env"
directory, which is rare enough to be commendable.
> Short summary, AFAICT:
> - btrfs, ext4 are the overall winners
> - xfs to, but creating/deleting many files was *very* slow
Maybe, and these conclusions are sort of plausible (but I prefer
JFS and XFS for different reasons); however they are not supported
by your results as they seem to me to lack much meaning, as what is
being measured is far from clear, and in particular it does not
seem to be the file system performance, or anyhow an aspect of
filesystem performance that might relate to common usage.
I think that it is rather better to run a few simple operations
(like the "generic" test) properly (unlike the "generic" test), to
give a feel for how well implemented are the basic operations of
the file system design.
Profiling a file system performance with a meaningful full scale
benchmark is a rather difficult task requiring great intellectual
fortitude and lots of time.
> - if you need only fast but no cool features or
> journaling, ext2 is still a good choice :)
That is however a generally valid conclusion, but with a very,
very important qualification: for freshly loaded filesystems.
Also with several other important qualifications, but "freshly
loaded" is a pet peeve of mine :-).
------------------------------------------------------------------------------
This SF.Net email is sponsored by the Verizon Developer Community
Take advantage of Verizon's best-in-class app development support
A streamlined, 14 day to market process makes app distribution fast and easy
Join now and get one step closer to millions of Verizon customers
http://p.sf.net/sfu/verizon-dev2dev
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: benchmark results
[not found] ` <4B336289.3010608-AanptEQQ3TL9uQeqpI+JUg@public.gmane.org>
@ 2009-12-24 15:35 ` Ryusuke Konishi
0 siblings, 0 replies; 16+ messages in thread
From: Ryusuke Konishi @ 2009-12-24 15:35 UTC (permalink / raw)
To: lists-AanptEQQ3TL9uQeqpI+JUg
Cc: linux-nilfs-u79uwXL29TY76Z2rM5mHXA, users-JrjvKiOkagjYtjvyW6yDsg,
konishi.ryusuke-Zyj7fXuS5i5L9jVzuh4AOg
On Thu, 24 Dec 2009 04:46:01 -0800, Christian Kujau wrote:
> On Thu, 24 Dec 2009 04:21:13 -0800 (PST), Christian Kujau wrote:
> > On Thu, 24 Dec 2009 at 21:06, Ryusuke Konishi wrote:
> > > > Bonnie++ couldn't complete on nilfs2, only the generic tests
> > > > and tiobench were run.
> > >
> > > I looked at the log but couldn't identify the error.
> > > Is that a disk full?
> >
> > Well, bonnie++ just exits with:
> >
> > Create files in sequential order...Can't write data.
> > Cleaning up test directory after error.
> >
> > ...which usually means that the disk is full. And I seem to remember that
> > nilfs2 filesystems were indeed filling up as data is (re)written. However,
> > I was just curious to test nilfs2 - I'm not even sure things like bonnie++
> > are supposed to run at all on a versioning file system :)
> >
> > I'll try to run a few more nilfs2 tests to find out of it's really an
> > -ENOSPACE thing.
>
> I was curious enough to just do that right now:
>
> $ cat t-nilfs2.sh
> #!/bin/sh
> umount /mnt/d1
> mkfs.nilfs2 /dev/xvdb
> mount -t nilfs2 -o barrier=off /dev/xvdb /mnt/d1
> for i in `seq 1 1000`; do
> date
> dd if=/dev/zero of=/mnt/d1/file.$i bs=1M count=100 2>/dev/null
> sync && rm -v /mnt/d1/file.$i && sync
> ls -la /mnt/d1
> df -k /mnt/d1 | grep -v Files
> echo
> done
> --------------
>
> So, basically I'm just writing 100MB to nilfs, then deleting the file
> again, sync and making sure that file.$i has been indeed deleted. Here's
> just the output from the df(1) commands:
>
> Filesystem 1K-blocks Used Available Use% Mounted on
> /dev/xvdb 2088956 114684 1867776 6% /mnt/d1
> /dev/xvdb 2088956 221180 1761280 12% /mnt/d1
> /dev/xvdb 2088956 327676 1654784 17% /mnt/d1
> /dev/xvdb 2088956 434172 1548288 22% /mnt/d1
> /dev/xvdb 2088956 524284 1458176 27% /mnt/d1
> /dev/xvdb 2088956 638972 1343488 33% /mnt/d1
> /dev/xvdb 2088956 745468 1236992 38% /mnt/d1
> /dev/xvdb 2088956 835580 1146880 43% /mnt/d1
> /dev/xvdb 2088956 958460 1024000 49% /mnt/d1
> /dev/xvdb 2088956 1056764 925696 54% /mnt/d1
> /dev/xvdb 2088956 1163260 819200 59% /mnt/d1
> /dev/xvdb 2088956 1261564 720896 64% /mnt/d1
> /dev/xvdb 2088956 1368060 614400 70% /mnt/d1
> /dev/xvdb 2088956 1482748 499712 75% /mnt/d1
> /dev/xvdb 2088956 1572860 409600 80% /mnt/d1
> /dev/xvdb 2088956 1687548 294912 86% /mnt/d1
> /dev/xvdb 2088956 1794044 188416 91% /mnt/d1
> /dev/xvdb 2088956 1892348 90112 96% /mnt/d1
> /dev/xvdb 2088956 1957884 24576 99% /mnt/d1
> /dev/xvdb 2088956 1974268 8192 100% /mnt/d1
> /dev/xvdb 2088956 1974268 8192 100% /mnt/d1
> /dev/xvdb 2088956 1974268 8192 100% /mnt/d1
> [...]
>
> With 8K free, dd(1) was unable to write any data to the filesystem. This
> is all with Linux 2.6.32 (x86-64) and nilfs2-tools 2.0.14-5, but I think
> this happened with earlier kernels too. Again, perhaps nilfs2 is
> supposed to work that way, somewhere the "old" versions of the
> filesystem have to be stored and this the fs fills up of course - but
> you tell me :-)
>
> Christian.
Thank you for the quick responses.
Yes, nilfs -- more precisely garbage collector of nilfs, by its
nature, preserves old data within a period of time. So, it usually
needs some care for this kind of test. Benchmark programs make fair
amount of changes for the short run, by its very nature ;)
Anyway, I'm relieved for now.
Thanks,
Ryusuke Konishi
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: benchmark results
2009-12-24 12:59 ` Teran McKinney
(?)
(?)
@ 2009-12-24 20:01 ` Christian Kujau
-1 siblings, 0 replies; 16+ messages in thread
From: Christian Kujau @ 2009-12-24 20:01 UTC (permalink / raw)
To: Teran McKinney; +Cc: reiserfs-devel, linux-ext4
[cut a few Cc's]
On Thu, 24 Dec 2009 at 12:59, Teran McKinney wrote:
> Which I/O scheduler are you using? Pretty sure that ReiserFS is a
> little less deadlocky with CFQ or another over deadline, but that
> deadline usually gives the best results for me (especially for JFS).
Yes, I'm using the deadline scheduler - it was my understanding that it's
usually best when I/O throughput is an issue. If reiserfs is "less
deadlocky" with other schedulers, maybe it should depend on
!IOSCHED_DEADLINE then or something :)
Christian.
--
BOFH excuse #19:
floating point processor overflow
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: benchmark results
2009-12-24 10:31 ` Christian Kujau
(?)
@ 2009-12-29 11:27 ` Emmanuel Florac
-1 siblings, 0 replies; 16+ messages in thread
From: Emmanuel Florac @ 2009-12-29 11:27 UTC (permalink / raw)
To: Christian Kujau
Cc: xfs, reiserfs-devel, linux-ext4, linux-btrfs, jfs-discussion,
ext-users, linux-nilfs
Le Thu, 24 Dec 2009 02:31:10 -0800 (PST) vous =E9criviez:
> - btrfs, ext4 are the overall winners
> - xfs to, but creating/deleting many files was *very* slow
xfs is slow at file creation/deletion if you mount it with barriers on
standard disk drives (not so with hardware RAID controllers).
--=20
-----------------------------------------------------------------------=
-
Emmanuel Florac | Direction technique
| Intellique
| <eflorac@intellique.com>
| +33 1 78 94 84 02
-----------------------------------------------------------------------=
-
--
To unsubscribe from this list: send the line "unsubscribe reiserfs-deve=
l" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: benchmark results
@ 2009-12-29 11:27 ` Emmanuel Florac
0 siblings, 0 replies; 16+ messages in thread
From: Emmanuel Florac @ 2009-12-29 11:27 UTC (permalink / raw)
To: Christian Kujau
Cc: xfs, reiserfs-devel, linux-ext4, linux-btrfs, jfs-discussion,
ext-users, linux-nilfs
Le Thu, 24 Dec 2009 02:31:10 -0800 (PST) vous écriviez:
> - btrfs, ext4 are the overall winners
> - xfs to, but creating/deleting many files was *very* slow
xfs is slow at file creation/deletion if you mount it with barriers on
standard disk drives (not so with hardware RAID controllers).
--
------------------------------------------------------------------------
Emmanuel Florac | Direction technique
| Intellique
| <eflorac@intellique.com>
| +33 1 78 94 84 02
------------------------------------------------------------------------
--
To unsubscribe from this list: send the line "unsubscribe reiserfs-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: benchmark results
@ 2009-12-29 11:27 ` Emmanuel Florac
0 siblings, 0 replies; 16+ messages in thread
From: Emmanuel Florac @ 2009-12-29 11:27 UTC (permalink / raw)
To: Christian Kujau
Cc: jfs-discussion, linux-nilfs, reiserfs-devel, xfs, ext-users,
linux-ext4, linux-btrfs
Le Thu, 24 Dec 2009 02:31:10 -0800 (PST) vous écriviez:
> - btrfs, ext4 are the overall winners
> - xfs to, but creating/deleting many files was *very* slow
xfs is slow at file creation/deletion if you mount it with barriers on
standard disk drives (not so with hardware RAID controllers).
--
------------------------------------------------------------------------
Emmanuel Florac | Direction technique
| Intellique
| <eflorac@intellique.com>
| +33 1 78 94 84 02
------------------------------------------------------------------------
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply [flat|nested] 16+ messages in thread
end of thread, other threads:[~2009-12-29 11:27 UTC | newest]
Thread overview: 16+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2008-08-03 16:43 Benchmark results bbmailing
-- strict thread matches above, loose matches on Subject: below --
2008-08-03 16:51 bbmailing
2008-08-03 17:07 ` Keir Fraser
2009-12-24 10:31 benchmark results Christian Kujau
2009-12-24 10:31 ` Christian Kujau
2009-12-24 12:06 ` Ryusuke Konishi
2009-12-24 12:06 ` Ryusuke Konishi
[not found] ` <alpine.DEB.2.01.0912240415450.3483@bogon.housecafe.de>
[not found] ` <4B336289.3010608@nerdbynature.de>
[not found] ` <4B336289.3010608-AanptEQQ3TL9uQeqpI+JUg@public.gmane.org>
2009-12-24 15:35 ` Ryusuke Konishi
2009-12-24 12:59 ` Teran McKinney
2009-12-24 12:59 ` Teran McKinney
2009-12-24 12:59 ` Teran McKinney
2009-12-24 20:01 ` Christian Kujau
2009-12-24 13:05 ` Peter Grandi
2009-12-29 11:27 ` Emmanuel Florac
2009-12-29 11:27 ` Emmanuel Florac
2009-12-29 11:27 ` Emmanuel Florac
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.