• <Intel.com 還有更多

Intel® Xeon Phi™ Product Family

Highly parallel processing to power your breakthrough innovations

Intel® Xeon Phi™ Coprocessor

Power your breakthrough innovations with the highly parallel processing of the Intel® Xeon Phi™ coprocessor. We have packed over a teraFLOPS of double-precision peak performance into every chip.

Life science applications

Life sciences is one of the largest commercial market segments for high-performance computing. Life sciences includes many disciplines, such as chemistry, bio-chemistry, molecular modeling, protein folding, genomics, and statistical analysis to name a few.  Quick simulation and analysis leads to breakthroughs in all areas of the life sciences. BWA maps human genomes. Gromacs simulates particle molecular dynamics. LAMMPS is a versatile large scale atomic simulation package. MPI-HMMER is an open source MPI implementation of the HMMER protein sequence analysis suite. NAMD is a parallel molecular dynamics code for large bio molecular systems. 

Intel measured as of May 2014

詳細配置

BWA: (Burrow-Wheeler Aligner Release bwa-0.5.10)

Platform hosting the coprocessor and platform for two-socket Intel® Xeon® processor baseline:

Two-socket Intel® Software Development Platform: 2x Intel Xeon processor E5-2697 v2 (12 core, 30M cache, 2.7 GHz, 8.0 GT/s Intel® QuickPath Interconnect (Intel® QPI), 130W thermal design power (TDP), Intel® Turbo Boost Technology on, Intel® Hyper-Threading Technology (Intel® HT Technology) on) 64 GB memory at 1600 MHz, Red Hat Enterprise Linux* (RHEL*) 6.2

Coprocessor details:

Intel® Xeon Phi™ coprocessor 7120A: 61 cores, 1.238 GHz, 16 memory channels, 16 GB memory at 5.5 GT/s, 300W TDP C-step (Intel® Turbo Boost Technology off, error-correcting code (ECC) on)

Software stack (Intel Xeon Phi coprocessor):

Intel® Manycore Platform Software Stack (Intel® MPSS) 2.1.6720-21 (Flash*: 2.1.03.0386; coprocessor OS: 2.6.38.8-gefd324e)

Intel® Composer XE 13.3.163.0, Intel® MPI Library 4.1.1.030

Two-socket Intel Xeon processor score: 178.73 seconds

Two-socket Intel Xeon processor + Intel Xeon Phi coprocessor score: 119.17 seconds

Source: Intel Internal Testing TR2093

 

MPIHmmer:

Platform hosting the coprocessor and platform for two-socket Intel® Xeon® processor baseline:

Two-socket Intel Software Development Platform: 2x Intel Xeon processor E5-2670 (8 core, 20M cache, 2.6 GHz, 8.0 GT/s Intel QPI, 115W TDP, Intel Turbo Boost Technology on, Intel HT Technology on) 64 GB memory at 1600 MHz, RHEL 6.4

Coprocessor details:

Intel® Xeon Phi™ coprocessor 7120A: 61 cores, 1.238 GHz, 16 memory channels, 16 GB memory at 5.5 GT/s, 300W TDP C-step (Intel Turbo Boost Technology off, ECC on)

Software stack (Intel Xeon Phi coprocessor):

Intel MPSS 2.1.6720-13 (Flash: 2.1.02.0386; coprocessor OS: 2.6.38.8-g5f2543d)

Intel® Composer XE 2013.3.163

Two-socket Intel Xeon processor score: 89 seconds

Two-socket Intel Xeon processor + Intel Xeon Phi coprocessor score: 57 seconds

Source: Intel Internal Testing TR2051

 

LAMMPS (Production Protein Simulation 474K Stoms, Version 5 April 2014)

Platform hosting the coprocessor and platform for two-socket Intel® Xeon® processor baseline:

Two-socket Intel Software Development Platform: 2x Intel Xeon processor E5-2697 v2 (2x 12 core, 30M cache, 2.7 GHz, 8.0 GT/s Intel QPI, 135W TDP, Intel Turbo Boost Technology on, Intel HT Technology on) 64 GB memory at 1600 MHz, RHEL 6.3

Coprocessor details:

Intel® Xeon Phi™ coprocessor 7120A: 61 cores, 1.238 GHz, 16 memory channels, 16 GB memory at 5.5 GT/s, 300W TDP C-step (Intel Turbo Boost Technology off, ECC on)

Software stack (Intel Xeon Phi coprocessor):

Intel MPSS 2.1.6720-13 (Flash: 2.1.02.0386)

Intel® Compiler 14.0.1; Intel MPI Library 4.1.2.040

 

1 Node

16 Nodes

Two-socket Intel Xeon processor score:

26.84 sec

4.33 seconds

Two-socket Intel Xeon processor + Intel Xeon Phi coprocessor score:

19.10 sec

2.63 seconds

Source: Intel Internal Testing TR2104

 

GROMACS  (Version 5.0-rc1 - Workload 512K H20 with RF Method)

Platform hosting the coprocessor and platform for two-socket Intel® Xeon® processor baseline:

Two-socket Intel Software Development Platform: 2x Intel Xeon processor E5-2697 v2 (12 core, 30M cache, 2.7 GHz, 8.0 GT/s Intel QPI, 130W TDP, Intel Turbo Boost Technology on, Intel HT Technology on) 64 GB memory at 1600 MHz, RHEL 6.4

Coprocessor details:

Intel® Xeon Phi™ coprocessor 7120A: 61 cores, 1.238 GHz, 16 memory channels, 16 GB memory at 5.5 GT/s, 300W TDP C-step (Intel Turbo Boost Technology off, ECC on)

Software stack (Intel Xeon Phi coprocessor):

Intel MPSS 2.1.6720-13 (Flash: 2.1.03.0386; coprocessor OS: 2.6.38.8-gefd324e)

Intel® Composer XE 13.3.163.0, Intel MPI Library 4.1.1.030

Two-socket Intel Xeon processor score: 2.312 ns/day

Two-socket Intel Xeon processor + Intel Xeon Phi coprocessor score: 3.605 ns/day seconds

Source: Intel Internal Testing TR2101

NAMD: (STMV 2.10 pre-release)

Platform hosting the coprocessor and platform for 2S Intel® Xeon® processor baseline:

W2600CR2 platform: 2x Intel Xeon processor E5-2697v2 (12 core, 30M cache, 2.7 GHz, 8.0 GT/s Intel QPI, 130W TDP, Intel Turbo Boost Technology on, Intel HT Technology on)  64 GB memory @ 1600 MHz, RHEL 6.2

Coprocessor details:

Intel® Xeon Phi™ coprocessor 7120A: 61 cores, 1.238 GHz, 16 memory channels, 16 GB memory @ 5.5 GT/s, 300W TDP C-step (Intel Turbo Boost Technology off, ECC on)

Software stack (Xeon Phi):

MPSS 2.1.6720-21 (Flash: 2.1.03.0386; coprocessor OS: 2.6.38.8-gefd324e)

Intel Compiler 13.1.3 20130607

2S Intel Xeon score: 0.513 seconds

2S Intel Xeon + 2 Xeon Phi score: 0.158 seconds

Source: Internal Testing (TR 2100)

 

NAMD STMV 1 node 47 PPN per node

(Version 2.10 pre-release)

STMV (virus) benchmark (1,066,628 atoms, periodic, PME)

Platform hosting the coprocessor:

W2600CR2 platform: 2x Intel Xeon processor E5-2697v2 (12 core, 30M cache, 2.7 GHz, 8.0 GT/s Intel QPI, 130W TDP)  64 GB memory @ 1600 MHz, RHEL 6.4 HT enabled

Intel® Xeon Phi™ coprocessor 7120A: 61 cores, 1.238 GHz, 16 memory channels, 16 GB memory @ 5.5 GT/s, 300W TDP C-step (ECC on, Intel Turbo Boost Technology off)

NVIDIA K40c GPU, 2880 SP cores, 960 DP cores, 745 MHz (nominal) 875 MHz boost, 12 memory channels, 12 GB memory @ 6.0 GT/s, ECC on, 235W TDP

CUDA 5.5.0, Driver 319.82 (boost frequency was not enabled for result)

Software stack:

MPSS 2.1.6720-16 (Flash 1.2.03.0386)

Intel Compiler 13.1.3 20130607 (2013.5.192)

1 Node

2 Nodes

4 Nodes

 

2S Xeon only score:

0.158 ns/day

     

2S Xeon + 1 Xeon Phi score:

0.386 ns/day

0.729 ns/day

1.2879 ns/day

(higher is better)

2S Xeon + 2 Xeon Phi score:

0.512 ns/day

     

NVIDIA K40c* score:

0.32 ns/day

0.55 ns/day

1.02 ns/day

(higher is better)

Fabric: 36 port switch/rack Mellanox FDR* (model MSX6025F-1BFR) Firmware version: 9_2_4002

Source: Internal Testing (TR 2100)

 

NAMD ApoA1 1 node (47 PPN per node)

Version 2.10 pre-release

Platform hosting the coprocessor:

W2600CR2 platform: 2x Intel Xeon processor E5-2697v2 (12 core, 30M cache, 2.7 GHz, 8.0 GT/s Intel QPI, 130W TDP) 64 GB memory @ 1600 MHz, RHEL 6.4 HT enabled

Intel® Xeon Phi™ coprocessor 7120A: 61 cores, 1.238 GHz, 16 memory channels, 16 GB memory @ 5.5 GT/s, 300W TDP C-step (ECC on, Turbo off)

NVIDIA K40c GPU, 2880 SP cores, 960 DP cores, 745 MHz (nominal) 875 MHz boost, 12 memory channels, 12 GB memory @ 6.0GT/s, ECC on, 235W TDP

CUDA 5.5.0, Driver 319.82(boost frequency was not enabled for result)

Software Stack:

MPSS 2.1.6720-16 (Flash 1.2.03.0386)

Intel Compiler 13.1.3 20130607 (2013.5.192)

1 Node

2 Nodes

 

2S Xeon + 1 Xeon Phi score (symmetric):

4.430 Mrps

6.607 Mrps

(higher is better)

NVIDIA K40c score:

4.08 Mrps

6.16 Mrps

(higher is better)

Fabric: 36 port switch/rack Mellanox FDR (model MSX6025F-1BFR) Firmware version: 9_2_4002

Source: Internal Testing (TR 2100)

Additional information: 1 2 3 4 5

產品與效能資訊

open

1. Software and workloads used in performance tests may have been optimized for performance only on Intel® microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations, and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information, go to www.intel.com/performance.

2. 對於本文件提及的各項第三方效能標竿或網站,Intel 並不會控制或稽核其設計或執行。Intel 鼓勵所有的客戶參閱提及的網站,或是其他發佈類似評量效能標竿報告的網站,確認提及的效能標竿是否正確,並且反映出可購買系統的效能表現。

3. Intel 處理器編號並不代表效能。處理器編號乃是用來區別屬於同一處理器系列中的不同特色,並非用來區別不同的處理器系列。如需詳細說明,請參閱 www.intel.com/content/www/us/en/processors/processor-numbers.html。

4. 對於不是 Intel® 微處理器特有的最佳化,用於非 Intel 微處理器時,Intel 的編譯器可能會,也可能不會最佳化到同樣程度。這些最佳化包括 SSE2 與 SSE3 指令集,也包括其他最佳化。對於任何最佳化,用於並非由 Intel 製造的微處理器,Intel 不保證最佳化的可用性、功能性或效力。 在本產品中取決於微處理器的最佳化,乃是為了用於 Intel 微處理器而設計。某些最佳化並非專門針對 Intel® 微架構,而是保留給 Intel 微處理器。關於本公告所涉及的具體指令集,如需詳細資訊,請參閱適用產品的使用指南與參考指南。 公告修訂版編號 20110804

5. 不同的硬體架構,可能需要不同的原始程式碼。其結果乃是根據 Intel 盡最大努力,使用經過最佳化,可在所有架構執行,並且執行相同工作的程式碼。日後的各種程式碼最佳化,可能會導致不同的結果。 在本產品中,取決於微處理器的最佳化,乃是為了用於 Intel® 微處理器而設計。某些最佳化並非專門針對 Intel® 微架構,而是保留給 Intel 微處理器。關於本公告所涉及的具體指令集,如需詳細資訊,請參閱適用產品的使用指南與參考指南。 公告修訂版編號 20110804