Linux Topology Benchmarks - dariof, jgross

	A	B	C	D	E	F	G	H	I	J	K	L	M	N	O	P
1	Intel(R) Xeon(R) X5650 @ 2.67GHz
2	pCPUs	48		DOM0
3	RAM	393138 MB		vCPUs	48
4	NUMA nodes	4		Memory	9955 MB
5
6	HACKBENCH	https://wiki.archlinux.org/index.php/Realtime_kernel#hackbench
7	# of task groups	"-g1”			“-g12”			“-g30”			“-g52”
8	vanilla/patched	vanilla	dariof	jgross	vanilla	dariof	jgorss	vanilla	dariof	jgross	vanilla	dariof	jgross
9		0.382	0.526	0.487	2.488	2.17	2.026	3.809	3.271	3.17	6.351	4.259	3.947
10		0.402	0.412	0.499	2.623	2.213	2.138	3.594	3.288	2.877	6.309	4.026	3.4
11	(lower == better)	0.418	0.397	0.556	2.445	2.3	2.085	3.819	3.522	3.108	6.043	4.369	3.72
12		0.387	0.542	0.501	2.611	2.378	2.076	3.151	2.944	2.688	5.043	3.741	3.89
13		0.43	0.554	0.513	2.605	2.086	2.392	3.499	3.242	2.909	5.683	3.671	3.697
14		0.398	0.469	0.518	2.189	2.283	2.365	3.81	2.926	3.26	6.28	3.415	3.812
15		0.421	0.444	0.518	2.526	2.196	2.146	3.502	3.655	3.341	6.052	4.43	3.989
16		0.394	0.473	0.513	2.525	2.126	2.364	3.136	3.178	3.498	5.614	3.23	3.825
17		0.414	0.492	0.532	2.561	2.283	2.326	3.847	3.51	2.898	6.123	3.943	3.764
18		0.396	0.478	0.456	2.615	2.167	2.114	3.552	2.713	3.153	5.81	3.62	3.409
19	Avg.	0.4042	0.4787	0.5093	2.5188	2.2202	2.2032	3.5719	3.2249	3.0902	5.9308	3.8704	3.7453
20	Std. Dev.	0.016	0.052	0.027	0.13	0.089	0.141	0.264	0.297	0.246	0.404	0.405	0.202
21	% increase vs vanilla		-18.431	-26.002		11.855	12.53		9.715	13.486		34.741	36.85
22
23	MAKE XEN
24	# of build jobs	“-j1”			“-j20”			“-j24”			“-j48”			“-j62”
25	vanilla/patched	vanilla	dariof	jgross	vanilla	dariof	jgross	vanilla	dariof	jgross	vanilla	dariof	jgross	vanilla	dariof	jgross
26		275.08	241.09	237.1	37.19	41.18	36.54	37.91	36.89	37.02	34.62	32.02	33.91	35.39	33.99	33.4
27		274.11	231.53	234.54	37.59	36.09	36.59	41.25	39.79	36.16	35.29	33.76	36.84	39.82	33.51	38.31
28	(lower == better)	272.93	235.2	234.16	37.1	36.42	36.59	37.44	35.9	35.73	34.69	37.18	34.7	34.66	33.04	34.28
29		270.79	236.79	239.8	38	39.97	37.17	37.59	35.81	35.16	35.59	33.1	32.62	34.26	37	37.1
30		272.92	238.67	233.59	37.27	36.05	36.85	38.18	40.1	34.99	34.68	34.46	32.57	34.84	33.07	33.87
31		272.62	234.42	235.15	37.82	36.54	39.92	37.27	36	35.98	35.27	32.99	33.63	34.61	33.39	37.9
32		269.34	229.82	238.37	38.2	37.14	36.58	42.25	35.55	35.26	34.99	32.85	33.81	39.78	33.06	33.33
33		270.52	239.13	233.89	37.23	36.36	36.29	41.4	35.44	35.21	35.36	33.24	34.08	34.13	32.41	33.76
34		269.96	235.38	237.84	37.98	35.89	36.06	38.1	35.04	41.1	34.79	33.01	39.29	39.82	37.27	33.09
35		269.63	232.91	230.96	37.53	36.66	36.4	37.55	39.85	39.41	35.39	33.44	33.16	35.18	33.07	33.58
36	Avg.	271.79	235.494	235.54	37.591	37.23	36.899	38.894	37.037	36.602	35.067	33.605	34.461	36.249	33.981	34.862
37	Std. Dev.	2.003	3.53	2.676	0.392	1.82	1.103	1.928	2.042	2.056	0.354	1.405	2.089	2.483	1.712	2.053
38	% increase vs vanilla		13.354	13.338		0.96	1.841		4.775	5.893		4.169	1.728		6.257	3.826
39
40	UNIXBENCH	https://code.google.com/p/byte-unixbench/
41	# parallel copies	1 parallel copy			20 parrallel copies			24 parallel copies			48 parallel copies			62 parallel copies
42	vanilla/patched	vanilla	dariof	jgross	vanilla	dariof	jgross	vanilla	dariof	jgross	vanilla	dariof	jgross	vanilla	dariof	jgross
43	Dhrystone 2 using register variables	2026.1	2021.3	2031	39840.8	39307.7	39237.2	44345.5	44200.2	44014.9	51402	51035.1	51386.7	51351	51977.5	51968.6
44	Double-Precision Whetstone	524.7	525.2	525.4	10385.2	10405.9	10357	12229	12199	12257.1	20948.6	20951.8	20938.8	26977.3	27102.2	27084
45	Execl Throughput	114.4	114.9	115.6	750.1	744.7	745.7	671.7	665.3	673.7	708.4	716.7	716.8	717.2	724.2	728.1
46	File Copy 1024 bufsize 2000 maxblocks	605.9	613	619.7	603.9	609.4	612	575.8	588.5	610.3	512.9	558.3	534.8	357.1	378.3	397.9
47	File Copy 256 bufsize 500 maxblocks	377.9	384	379.3	430.8	424.3	411.3	385.1	417.5	423.8	371.8	389.2	368.1	230.3	262	248.2
48	File Copy 4096 bufsize 8000 maxblocks	1127	932.6	1140.6	1298	1270.8	1265.4	1229.9	1265.5	1229.5	1098.7	1147.3	1167.2	798.9	947.6	880
49	Pipe Throughput	285.4	286.2	284.3	5638.7	5612.4	5626.4	6563.5	6658.1	6596.2	10353	10432.3	10441.2	10503	10538.8	10456.2
50	Pipe-based Context Switching	51.5	143.7	151	515.4	1886.6	1939.6	699.7	2065.9	2109.4	2746.5	3172.8	3207.9	2905.5	3229.8	3313.1
51	Process Creation	61.2	81.5	82.4	364	376	372.7	330.1	345.8	348.3	364.3	371.8	373.8	365.2	356.6	356.9
52	Shell Scripts (1 concurrent)	443.9	480.9	481.5	1829.9	1890.3	1890.3	1693.1	1732.4	1736.4	2308.3	2294.9	2302	2259.9	2271.2	2268.1
53	Shell Scripts (8 concurrent)	1248.5	1305.8	1307	2161.6	2175.7	2177.7	2151.5	2161.2	2163.8	2101.5	2098.1	2103.8	2105.7	2098.2	2091.4
54	System Call Overhead	256.9	254	255.7	892.9	894	906.6	972	970.2	983.1	1455.8	1450.1	1478.7	1477.6	1470	1500.8
55	System Benchmarks Index Score	355.1	394.8	403.9	1644.6	1835.8	1835.8	1687.5	1878.2	1889.4	2182.7	2245.3	2239.2	2037.4	2118	2111.2
56	% increase (of Index Score) vs vanilla		11.18	13.743		11.626	11.626		11.301	11.964		2.868	2.589		3.956	3.622
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100