2. Click to edit Master title style
2
Introduction
High-level overview of the Intel® Xeon Phi™ platform:
Hardware and Software
Performance and Thread Parallelism
Conclusions & References
5. Click to edit Master title style
5
Introduction
High-level overview of the Intel® Xeon Phi™ platform:
Hardware and Software
Performance and Thread Parallelism
Conclusions & References
12. Click to edit Master title style
12
Introduction
High-level overview of the Intel® Xeon Phi™ platform:
Hardware and Software
Performance Considerations
Performance and Thread Parallelism
Conclusions & References
17. INTEL CONFIDENTIAL17
75
171
0
50
100
150
200
STREAM
Triad (GB/s)
330
802
0
200
400
600
800
1000
SMP Linpack
(GF/s)
347
887
0
200
400
600
800
1000
DGEMM
(GF/s)
728
1,796
0
500
1000
1500
2000
SGEMM
(GF/s)
Notes
1. Intel® Xeon® Processor E5-2680 used for all SGEMM Matrix = 12800 x 12800 , DGEMM Matrix 10752 x
10752, SMP Linpack Matrix 26000 x 26000
2. Intel® Xeon Phi™ coprocessor SE10P (ECC on) with “Gold” SW stack SGEMM Matrix = 12800 x 12800,
DGEMM Matrix 12800 x 12800, SMP Linpack Matrix 26872 x 28672
3. Average single-node results from measurements across a set of nodes from the TACC+ Stampede* Cluster
+ Texas Advanced Computing Center (TACC) at the University of Texas at Austin.
++ Measured on the TACC+ Stampede Cluster
Coprocessor results: Benchmark run 100% on coprocessor,
no help from Intel® Xeon® processor host (aka native)
Synthetic Benchmarks
Intel® Xeon Phi™ Coprocessor and Intel® MKL
UP TO
2.4X
UP TO
2.5X
UP TO
2.2X
UP TO
2.4X
Higher is Better
• 2S Intel® Xeon®
• Intel Xeon Phi
ECC ON84% Efficient 83% Efficient 75% Efficient
18. Click to edit Master title style
18
Introduction
High-level overview of the Intel® Xeon Phi™ platform:
Hardware and Software
Native, Offload and Variations
Performance and Thread Parallelism
Conclusions & References
39. Click to edit Master title style
39
Introduction
High-level overview of the Intel® Xeon Phi™ platform:
Hardware and Software
Performance and Thread Parallelism
Conclusions & References
42. Click to edit Master title style
42
Introduction
High-level overview of the Intel® Xeon Phi™ platform:
Hardware and Software
Performance and Thread Parallelism: OpenMP
Conclusions & References
47. Click to edit Master title style
47
Introduction
High-level overview of the Intel® Xeon Phi™ platform:
Hardware and Software
Performance and Thread Parallelism: MKL
Conclusions & References
53. Click to edit Master title style
53
Introduction
High-level overview of the Intel® Xeon Phi™ platform:
Hardware and Software
Performance and Thread Parallelism
Conclusions & References