PGI Compiler 18.10 Benchmarks vs. GCC vs. LLVM Clang PGI compiler benchmarks for a future article on Phoronix.com. PGI Compiler 18.10: Processor: Intel Core i9-7980XE @ 4.20GHz (18 Cores / 36 Threads), Motherboard: ASUS PRIME X299-A (1602 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 16384MB, Disk: 240GB Force MP510, Graphics: NVIDIA NV120 12GB, Audio: Realtek ALC1220, Monitor: ASUS PB278, Network: Intel Connection OS: Ubuntu 18.10, Kernel: 4.20.0-999-generic (x86_64) 20181206, Desktop: GNOME Shell 3.30.1, Display Server: X Server 1.20.1, Display Driver: modesetting 1.20.1, OpenGL: 4.3 Mesa 18.2.2, Compiler: PGI Compiler 18.10-1, File-System: ext4, Screen Resolution: 2560x1440 GCC 8.2.0: Processor: Intel Core i9-7980XE @ 4.20GHz (18 Cores / 36 Threads), Motherboard: ASUS PRIME X299-A (1602 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 16384MB, Disk: 240GB Force MP510, Graphics: NVIDIA NV120 12GB, Audio: Realtek ALC1220, Monitor: ASUS PB278, Network: Intel Connection OS: Ubuntu 18.10, Kernel: 4.20.0-999-generic (x86_64) 20181206, Desktop: GNOME Shell 3.30.1, Display Server: X Server 1.20.1, Display Driver: modesetting 1.20.1, OpenGL: 4.3 Mesa 18.2.2, Compiler: GCC 8.2.0, File-System: ext4, Screen Resolution: 2560x1440 LLVM Clang 7.0: Processor: Intel Core i9-7980XE @ 4.20GHz (18 Cores / 36 Threads), Motherboard: ASUS PRIME X299-A (1602 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 16384MB, Disk: 240GB Force MP510, Graphics: NVIDIA NV120 12GB, Audio: Realtek ALC1220, Monitor: ASUS PB278, Network: Intel Connection OS: Ubuntu 18.10, Kernel: 4.20.0-999-generic (x86_64) 20181206, Desktop: GNOME Shell 3.30.1, Display Server: X Server 1.20.1, Display Driver: modesetting 1.20.1, OpenGL: 4.3 Mesa 18.2.2, File-System: ext4, Screen Resolution: 2560x1440 BlogBench 1.1 Test: Read Final Score > Higher Is Better PGI Compiler 18.10 . 752310 |======================================= GCC 8.2.0 .......... 709382 |===================================== LLVM Clang 7.0 ..... 962631 |================================================== High Performance Conjugate Gradient 3.0 GFLOP/s > Higher Is Better PGI Compiler 18.10 . 1.34 |==================================================== GCC 8.2.0 .......... 1.34 |==================================================== LLVM Clang 7.0 ..... 1.33 |==================================================== PolyBench-C 4.2 Test: Covariance Computation Seconds < Lower Is Better PGI Compiler 18.10 . 7.23 |==================================================== GCC 8.2.0 .......... 4.86 |=================================== LLVM Clang 7.0 ..... 4.87 |=================================== PolyBench-C 4.2 Test: Correlation Computation Seconds < Lower Is Better PGI Compiler 18.10 . 4.81 |=================================================== GCC 8.2.0 .......... 4.86 |==================================================== LLVM Clang 7.0 ..... 4.88 |==================================================== PolyBench-C 4.2 Test: 3 Matrix Multiplications Seconds < Lower Is Better PGI Compiler 18.10 . 3.35 |==================================================== GCC 8.2.0 .......... 2.86 |============================================ LLVM Clang 7.0 ..... 2.83 |============================================ Timed HMMer Search 2.3.2 Pfam Database Search Seconds < Lower Is Better PGI Compiler 18.10 . 8.51 |============================================ GCC 8.2.0 .......... 9.99 |==================================================== LLVM Clang 7.0 ..... 8.28 |=========================================== SciMark 2.0 Computational Test: Composite Mflops > Higher Is Better PGI Compiler 18.10 . 1932 |======================================== GCC 8.2.0 .......... 2518 |==================================================== LLVM Clang 7.0 ..... 2478 |=================================================== SciMark 2.0 Computational Test: Monte Carlo Mflops > Higher Is Better PGI Compiler 18.10 . 591 |================================= GCC 8.2.0 .......... 950 |===================================================== LLVM Clang 7.0 ..... 717 |======================================== SciMark 2.0 Computational Test: Fast Fourier Transform Mflops > Higher Is Better PGI Compiler 18.10 . 768 |==================================================== GCC 8.2.0 .......... 773 |===================================================== LLVM Clang 7.0 ..... 778 |===================================================== SciMark 2.0 Computational Test: Sparse Matrix Multiply Mflops > Higher Is Better PGI Compiler 18.10 . 2243 |================================== GCC 8.2.0 .......... 3405 |==================================================== LLVM Clang 7.0 ..... 3164 |================================================ SciMark 2.0 Computational Test: Dense LU Matrix Factorization Mflops > Higher Is Better PGI Compiler 18.10 . 4622 |======================================== GCC 8.2.0 .......... 5796 |================================================== LLVM Clang 7.0 ..... 6069 |==================================================== SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation Mflops > Higher Is Better PGI Compiler 18.10 . 1438 |============================================= GCC 8.2.0 .......... 1665 |==================================================== LLVM Clang 7.0 ..... 1662 |==================================================== TSCP 1.81 AI Chess Performance Nodes Per Second > Higher Is Better PGI Compiler 18.10 . 1199430 |==================================== GCC 8.2.0 .......... 1440321 |=========================================== LLVM Clang 7.0 ..... 1629136 |================================================= C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel Seconds < Lower Is Better PGI Compiler 18.10 . 39.24 |=================== GCC 8.2.0 .......... 103.10 |================================================== LLVM Clang 7.0 ..... 52.79 |========================== AOBench Size: 2048 x 2048 - Total Time Seconds < Lower Is Better PGI Compiler 18.10 . 24.15 |==================================== GCC 8.2.0 .......... 32.18 |================================================= LLVM Clang 7.0 ..... 33.81 |===================================================