Benchmark Information

Competitive and PGI SPEC® OMPG2012 results listed below were conducted by PGI on the dates listed. Competitor results were conducted using what PGI believed to be the best known optimization options at the time the results were conducted.

SPEC benchmarks include two fundamental metrics. "Base" performance requires the use of a uniform set of compiler options and runtime settings across all benchmarks in the suite. "Peak" performance allows use of arbitrary combinations of compiler options, runtime settings and profile-feedback optimization to maximize performance of individual benchmarks in the suite. PGI believes Base is most representative of "compile-and-go" performance, and it is for this reason that the comparisons on these pages list 64-bit Base performance only.

PGI uses OMPG2012 primarily to compare compiler performance over time. PGI does not perform direct comparative system peformance testing. To minimize performance fluctuations due to hardware variations, PGI retains a limited set of reference systems whose configurations are fixed and stable. As a general rule, reference systems are not modified or upgraded; instead new systems are added and old systems are retired from the reference system pool as needed. The results posted on this web site are organized by individual reference systems. Higher scores are better.

PGI Reference System: "PFL-HW01" (32-core Intel Xeon)

Date Compiler SPECompG_base2012  Notes
Aug-2015 PGI 15.7 7.97 Two threads per core, Hyperthreading enabled
Apr-2015 PGI 15.4 7.92 Two threads per core, Hyperthreading enabled
Feb-2015 PGI 15.1 7.87 Two threads per core, Hyperthreading enabled
Feb-2015 Intel 15.0.2 8.02 Two threads per core, hyperthreading enabled

System Information:

SuperMicro 2028GR-TRT with two Intel Xeon E5-2698 v3 at 2.30GHz processors. The system has two 16-core chips and 256GB of memory (16X16 GB PC4-1490R-15 ECC). Operating System: CentOS 6.6.

PGI Reference System: "Galaxy" (16-core Intel Xeon)

Date Compiler SPECompG_base2012  Notes
Apr-2014 PGI 14.4 4.07 One thread per core, Hyperthreading enabled
Apr-2014 Intel 14.0.2 4.03 One thread per core, hyperthreading enabled
Jan-2014 PGI 14.1 3.99 One thread per core, Hyperthreading enabled
Jan-2014 Intel 14.0.1 4.02 One thread per core, hyperthreading enabled
Jul-2013 PGI 13.7 3.96 One thread per core, Hyperthreading enabled
Jul-2013 Intel 13.1.3 4.02 One thread per core, hyperthreading enabled
Jul-2013 GCC 4.8.1 2.23 One thread per core, hyperthreading enabled
Jan-2013 PGI 13.1 3.73 One thread per core
Jan-2013 Intel 13.0 3.80 One thread per core
Jan-2013 GCC 4.7 2.10 Estimates*

System Information:

HP ProLiant SL250 with two Intel Xeon E5-2670 at 2.6GHz processors. The system has two eight-core chips and 64GB of memory (8x8 GB 2Rx4 PC3-8500R ECC CL5). Operating System: RHEL 6.2 (64-bit) kernel version 2.6.32-220.e16.x86_64.

PGI Reference System: "Piledriver" (32-core AMD Opteron)

Date Compiler SPECompG_base2012  Notes
Apr-2014 PGI 14.4 4.32 One thread per core
Apr-2014 Intel 14.0.2 3.92 One thread per core
Jan-2014 PGI 14.1 4.29 One thread per core
Jan-2014 Intel 14.0.1 3.91 One thread per core
Jul-2013 PGI 13.7 4.25 One thread per core
Jul-2013 Intel 13.1.3 3.94 One thread per core
Jul-2013 GCC 4.8.1 2.64 One thread per core
Mar-2013 PGI 13.2 4.24 BIOS Performance defaults loaded
Jan-2013 PGI 13.1 3.52
Jan-2013 Intel 13.0 3.17
Jan-2013 GCC 4.7 2.32 Estimates*

System Information:

Supermicro A+ Server model 2022G-URF with AMD Opteron 6386 SE at 2.8GHz processors. The system has two 16-core chips and 64GB of memory (8x8 GB 2Rx4 PC3L-12800R-11 ECC). Operating System: RHEL 6.2 (64-bit) kernel version 2.6.32-220.e16.x86_64.

* Notes: GCC 4.7 was unable to compile and run all test. For calculation purposes, PGI results were substituted for the failing tests. Thererfore the results shown for GCC are estimates only.

SPEC® is a registered trademarks of the Standard Performance Evaluation Corporation (SPEC).

Click me