Abstract
Optimizing HPC systems based on performance factors and bottlenecks is essential for designing an HPC infrastructure with the best characteristics and at a reasonable cost. Such insight can only be achieved through a detailed analysis of existing HPC systems and the execution of their workloads. The “Quinde I” is the only and most powerful supercomputer in Ecuador and is currently listed third on the South America. It was built with the IBM Power 8 servers. In this work, we measured its performance using different parameters from High-Performance Computing (HPC) to compare it with theoretical values and values obtained from tests on similar models. To measure its performance, we compiled and ran different benchmarks with the specific optimization flags for Power 8 to get the maximum performance with the current configuration in the hardware installed by the vendor. The inputs of the benchmarks were varied to analyze their impact on the system performance. In addition, we compile and compare the performance of two algorithms for dense matrix multiplication SRUMMA and DGEMM.
Highlights
Today’s computational methods such as modeling and simulations are the core tools for finding solutions to various biological to engineering problems that can only be solved by an High-Performance Computing (HPC) platform
A benchmark is a program or set of programs executed on a single machine or cluster to obtain the maximum performance of a given function under certain conditions and to compare the performance results with the measured values of similar machines or the theoretically expected values
We demonstrate High-Performance Linpack (HPL) performance on our HPC “Quinde I”, which is equipped with high technology with the following features: a cluster with IBM Power 8 processors, 1760 cores, RAM 11 TB, parallel memory 350 TB, NVIDIA K80 Tesla graphics processing units (GPUs) and InfiniBand technology
Summary
Today’s computational methods such as modeling and simulations are the core tools for finding solutions to various biological to engineering problems that can only be solved by an HPC platform. Performance evaluation and analysis have been a core topic in HPC research. High-Performance Linpack (HPL) benchmark, used in the TOP500 list as a long-established standard for measuring computational performance, has been challenged in recent years [1,2]. The LINPACK (HPL) benchmark was the defacto metric for ranking HPC systems, measuring the sustained floating point rate (GFLOPs/s) for solving a dense system of linear equations using double precision floating point arithmetic. The use of HPCC in performance modeling and prediction has already been investigated in the following work: Refs. Pfeiffer et al [5]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
