arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.03357math.NAcs.NAcs.PF
收藏

混合精度龙格-库塔法的性能评价

Performance Evaluation of Mixed-Precision Runge-Kutta Methods

Ben Burnett, Sigal Gottlieb, Zachary J. Grant, Alfa Heryudono

摘要:

在文献[8]中提出并分析了为在混合精度计算中保持高精度解而设计的附加龙格-库塔法。这些特殊设计的方法使用降低的精度或隐式计算和全精度的显式计算。在IBM POWER 9和Intel x86 64芯片上开发了一个用MP-ARK方法求解非线性系统常微分方程的FORTRAN程序。这些方法的收敛性,准确性,运行时,和能源消耗进行了探讨。我们表明,这些MP-ARK方法有效地产生准确的解决方案,显着减少运行时间(和扩展的能源消耗)。

英文摘要:

Additive Runge-Kutta methods designed for preserving highly accurate solutions in mixed-precision computation were proposed and analyzed in [8]. These specially designed methods use reduced precision or the implicit computations and full precision for the explicit computations. We develop a FORTRAN code to solve a nonlinear system of ordinary differential equations using the mixed precision additive Runge-Kutta (MP-ARK) methods on IBM POWER9 and Intel x86\_64 chips. The convergence, accuracy, runtime, and energy consumption of these methods is explored. We show that these MP-ARK methods efficiently produce accurate solutions with significant reductions in runtime (and by extension energy consumption).

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑