scholarly journals Co-design of a Particle-in-Cell Plasma Simulation Code for Intel Xeon Phi: A First Look at Knights Landing

2016 ◽  
pp. 319-329 ◽  
Author(s):  
Igor Surmin ◽  
Sergey Bastrakov ◽  
Zakhar Matveev ◽  
Evgeny Efimenko ◽  
Arkady Gonoskov ◽  
...  
Keyword(s):  
Plasma Simulation ◽  
Xeon Phi ◽  
Intel Xeon Phi ◽  
Simulation Code ◽  
Particle In Cell ◽  
Cell Plasma ◽  
Knights Landing ◽  
Intel Xeon

scholarly journals Three-dimensional particle-in-cell plasma simulation on Intel Xeon Phi: performance optimization and case study

2015 ◽  
pp. 486-500
Author(s):  
И.Б. Мееров ◽  
С.И. Бастраков ◽  
И.А. Сурмин ◽  
А.А. Гоносков ◽  
Е.С. Ефименко ◽  
...  
Keyword(s):  
Performance Optimization ◽  
Three Dimensional ◽  
Plasma Simulation ◽  
Computational Time ◽  
Xeon Phi ◽  
Intel Xeon Phi ◽  
Simulation Code ◽  
Particle In Cell ◽  
Cell Plasma ◽  
Intel Xeon

Рассматривается проблема эффективного использования ускорителей Xeon Phi при моделировании лазерной плазмы. Приводится анализ особенностей архитектуры Xeon Phi, влияющих на производительность кода при численном моделировании плазмы методом частиц в ячейках. Описывается параллельный программный комплекс PICADOR, оптимизированный ранее для расчетов на ускорителях. Производительность программного комплекса на Xeon Phi в сравнении с CPU исследуется при решении трех вычислительно трудоемких задач. Обсуждается соотношение времени расчета на Xeon Phi и CPU на разных этапах метода частиц в ячейках. Демонстрируется, что в зависимости от особенностей задачи Xeon Phi может как опережать, так и отставать от CPU при выполнении расчетов. An efficient application of computational systems equipped with Intel Xeon Phi coprocessors for the laser-plasma simulation is considered. The features of Xeon Phi architecture that influence the performance of Particle-in-Cell plasma simulation are analyzed. The PICADOR parallel plasma simulation code previously optimized for Xeon Phi is described. Its performance on Xeon Phi compared to CPU is studied on three computationally intensive plasma simulation problems. The ratio of computational time on Xeon Phi to CPU is discussed for the main stages of the Particle-in-Cell method. It is shown that, depending on the features of a physical problem, the use of Xeon Phi can be both advantageous and disadvantageous compared to CPU.

scholarly journals MILC Code Performance on High End CPU and GPU Supercomputer Clusters

2018 ◽  
Vol 175 ◽  
pp. 02009
Author(s):  
Carleton DeTar ◽  
Steven Gottlieb ◽  
Ruizi Li ◽  
Doug Toussaint
Keyword(s):  
Conjugate Gradient ◽  
Memory Hierarchy ◽  
Xeon Phi ◽  
Intel Xeon Phi ◽  
Code Performance ◽  
Recent Developments ◽  
Knights Landing ◽  
Many Core ◽  
Intel Xeon

With recent developments in parallel supercomputing architecture, many core, multi-core, and GPU processors are now commonplace, resulting in more levels of parallelism, memory hierarchy, and programming complexity. It has been necessary to adapt the MILC code to these new processors starting with NVIDIA GPUs, and more recently, the Intel Xeon Phi processors. We report on our efforts to port and optimize our code for the Intel Knights Landing architecture. We consider performance of the MILC code with MPI and OpenMP, and optimizations with QOPQDP and QPhiX. For the latter approach, we concentrate on the staggered conjugate gradient and gauge force. We also consider performance on recent NVIDIA GPUs using the QUDA library.

Accelerating Seismic Simulations Using the Intel Xeon Phi Knights Landing Processor

2017 ◽  
pp. 139-157 ◽  
Author(s):  
Josh Tobin ◽  
Alexander Breuer ◽  
Alexander Heinecke ◽  
Charles Yount ◽  
Yifeng Cui
Keyword(s):  
Xeon Phi ◽  
Intel Xeon Phi ◽  
Knights Landing ◽  
Intel Xeon

scholarly journals DD-αAMG on QPACE 3

2018 ◽  
Vol 175 ◽  
pp. 02007 ◽  
Author(s):  
Peter Georg ◽  
Daniel Richtmann ◽  
Tilo Wettig
Keyword(s):  
First Generation ◽  
Xeon Phi ◽  
Intel Xeon Phi ◽  
Knights Landing ◽  
Speedup Factor ◽  
Single Processor ◽  
Intel Xeon

We describe our experience porting the Regensburg implementation of the DD-αAMG solver from QPACE 2 to QPACE 3. We first review how the code was ported from the first generation Intel Xeon Phi processor (Knights Corner) to its successor (Knights Landing). We then describe the modifications in the communication library necessitated by the switch from InfiniBand to Omni-Path. Finally, we present the performance of the code on a single processor as well as the scaling on many nodes, where in both cases the speedup factor is close to the theoretical expectations.

Sign in / Sign up

Export Citation Format

Share Document