Seismic Processing Performance Analysis on Different Hardware Environment
The dynamic behavior of chosen mini-applications was studied using BSC performance analysis tools to identify their common features.
In summary, we observe the best performance of mini-applications on Intel Xeon E5-2698 CPU generation 4. Intel Xeon Phi 7250 peculiar architectural characteristics requires careful source code optimizations to help the compiler to effectively vectorize time-consuming loops and to improve the cache locality in order to achieve higher performance level. Elbrus-4S CPU is theoretically suitable for such kind of applications, but currently observed performance is an order of magnitude less than on Xeon E5 family, we believe that the frequency and RAM bandwidth increasing, as well as source code optimization work could improve it’s performance.
Dimemas: predict parallel performance using a single cpu machine ; https://tools.bsc.es/dimemas (accessed: 04.07.2017)
Gazdag J. Wave Equation migration with equation migration with Phase Shift method / J. Gazdag // Geophysics - 1978 - V. 43 P. 1342-1351
The GEOLAB-IT Company; http://sk.ru/net/1120173/
Home: Extrae, Trace-generation package; https://tools.bsc.es/extrae (accessed: 04.07.2017)
Intel Hyper-Threading Technology, http://www.intel.ru/content/www/ru/ru/architecture-and-technology/
hyper-threading/hyper-threading-technology.html, (accessed: 29.11.2016)
Performance Analysis Tools: Details and Intelligence; https://tools.bsc.es/, (accessed: 04.07.2017)
Popovici A.M. Prestack migration by split-step DSR / A.M. Popovici // Geophysics - 1996 - V. 61 - P. 1412-1416
Intel 64 and IA-32 Architectures Software Developers Manual, Volume 3B: System Programming Guide, Part 2, http://www.intel.com/content/dam/www/public/us/en/documents/ manuals/64-ia-32-architectures-software-developer-vol-3b-part-2-manual.pdf