PGI 18.10 Compiler Benchmarks Against GCC 8.2, LLVM Clang 7.0

Given the recently release of the PGI 18.10 Community Edition compiler by NVIDIA, I was curious to see how the performance on the CPU is looking for this proprietary compiler on Linux. For those curious as well, here are some benchmarks of the PGI 18.10 C/C++ compiler against the GCC 8.2.0 and LLVM Clang 7.0 open-source compilers.

From an Intel Core i9 7980XE system running Ubuntu 18.10, I benchmarked the PGHI 18.10 compiler against GCC 8.2 and LLVM Clang 7.0 under a variety of C/C++ benchmarks to explore the performance of the resulting binaries. The PGI compiler also has great GPU offloading support with NVIDIA hardware, but for this initial 18.10 comparison was just exploring the CPU performance while maintaining the same CFLAGS/CXXFLAGS throughout testing.

All of these PGI/GCC/Clang compiler benchmarks were carried out using the Phoronix Test Suite.

First up was BlogBench where the binary produced by PGI 18.10 Community Edition was slightly faster than GCC 8.2.0 but slower than LLVM Clang 7.0.

In the case of PolyBench-C, the results were mixed from PGI 18.10 being slowest in two of the cases to having a very narrow lead in the correlation computation test.

With the HMMer profile hidden markov models benchmark, the PGI 18.10 performance was just behind LLVM Clang 7.0 for a second place finish.