Skip to content

Intel IvyBridge S2 M1 C10

Jump to bottom
moebiusband73 edited this page Jan 18, 2021 · 4 revisions

System

  • Processor: Intel(R) Xeon(R) CPU E5-2690 v2
  • Base frequency: 3.0 GHz
  • Number of sockets: 2
  • Number of memory domains per socket: 1
  • Memory domain specs: 4-channel DDR-1866
  • Number of cores per socket: 10
  • Number of HWThreads per core: 2
  • MachineState output: json

Tool chain

+----------+---------------------------------+
| Compiler | icc (ICC)                       |
|----------|---------------------------------|
| Version  |  icc (ICC) 19.0.5.281 20190815  |
+----------+---------------------------------+

Optimizing flags: -fast -xHost -qopt-streaming-stores=always -std=c99 -ffreestanding -qopenmp

Results

All results are in GB/s.

Summary results:

+--------------------------------------------+
| Single core   |  18.16 (Sum)               |
| Memory domain |  56.37 (Sum with 7 cores)  |
| Socket        |  56.37 (Sum with 7 cores)  |
| Node          |  112.52 (Sum with 8 cores) |
+--------------------------------------------+

Results for scaling within a memory domain:

#nt	Init	Sum	Copy	Update	Triad	Daxpy	STriad	SDaxpy
1	5.45	18.16	7.46	12.31	10.05	15.08	12.31	17.19
2	10.51	32.42	14.26	22.51	19.10	27.15	23.14	30.63
3	15.36	43.33	20.65	31.78	27.27	37.18	32.49	40.99
4	20.03	50.42	26.62	38.78	34.66	43.28	40.37	46.17
5	24.73	53.84	32.36	43.82	40.83	46.09	45.52	48.33
6	29.26	55.56	37.48	46.27	44.77	47.35	47.65	49.12
7	33.56	56.37	41.34	47.36	46.57	47.88	48.26	49.24
8	37.46	56.31	45.35	47.36	47.47	47.68	48.70	49.03
9	41.04	56.22	45.04	47.42	47.20	47.83	48.06	49.05
10	44.07	56.00	45.45	47.12	46.90	47.53	47.64	48.73

Results for scaling across memory domains. Shown are the results for the number of memory domains used (nm) with columns number of cores used per memory domain.

Init:

#nm	1	2
1	5.45	12.91
2	10.51	24.64
3	15.36	35.60
4	20.03	46.04
5	24.73	56.70
6	29.26	66.57
7	33.56	75.66
8	37.46	83.37
9	41.04	90.02
10	44.07	95.47

Sum:

#nm	1	2
1	18.16	36.33
2	32.42	63.90
3	43.33	87.07
4	50.42	99.89
5	53.84	107.22
6	55.56	110.58
7	56.37	112.30
8	56.31	112.52
9	56.22	112.31
10	56.00	111.59

Copy

#nm	1	2
1	7.46	17.09
2	14.26	32.11
3	20.65	46.09
4	26.62	58.69
5	32.36	71.01
6	37.48	80.67
7	41.34	86.94
8	45.35	90.19
9	45.04	91.66
10	45.45	92.02

Update

#nm	1	2
1	12.31	24.54
2	22.51	44.64
3	31.78	63.27
4	38.78	77.19
5	43.82	87.60
6	46.27	92.13
7	47.36	94.69
8	47.36	95.56
9	47.42	95.64
10	47.12	94.62

Triad

#nm	1	2
1	10.05	22.61
2	19.10	41.89
3	27.27	59.55
4	34.66	74.30
5	40.83	85.81
6	44.77	92.05
7	46.57	94.78
8	47.47	95.48
9	47.20	95.27
10	46.90	94.70

Scaling

Memory bandwidth scaling within one memory domain: Main memory bandwidth scaling plot

The following plots illustrate the the performance scaling over multiple memory domains using different number of cores per memory domain.

Memory bandwidth scaling across memory domains for init: Memory domain scaling plot

Memory bandwidth scaling across memory domains for sum Memory domain scaling plot

Memory bandwidth scaling across memory domains for copy Memory domain scaling plot

Memory bandwidth scaling across memory domains for Triad Memory domain scaling plot

Clone this wiki locally