A Divide and Conquer Strategy for Scaling Weather Simulations with Multiple Regions of Interest

Preeti Malakar,Rashmi Mittal,Vaibhav Saxena,Vijay Natarajan,Thomas George,Yogish Sabharwal,Sameer Kumar,Sathish S Vadhiyar

doi:10.1155/2013/682356

Abstract

Accurate and timely prediction of weather phenomena, such as hurricanes and flash floods, require high-fidelity compute intensive simulations of multiple finer regions of interest within a coarse simulation domain. Current weather applications execute these nested simulations sequentially using all the available processors, which is sub-optimal due to their sub-linear scalability. In this work, we present a strategy for parallel execution of multiple nested domain simulations based on partitioning the 2-D processor grid into disjoint rectangular regions associated with each domain. We propose a novel combination of performance prediction, processor allocation methods and topology-aware mapping of the regions on torus interconnects. Experiments on IBM Blue Gene systems using WRF show that the proposed strategies result in performance improvement of up to 33% with topology-oblivious mapping and up to additional 7% with topology-aware mapping over the default sequential strategy.

Highlights

Accurate and timely prediction of catastrophic events such as hurricanes, heat waves, and thunderstorms enables policy makers to take quick preventive actions
Experiments on IBM Blue Gene systems show that the proposed performance modeling, partitioning and processor allocation strategies can improve simulation performance over the default strategy of employing the maximum number of processors for all the nested simulations by up to 33% with topology-oblivious mapping and up to an additional 7% with topology-aware mapping
We show that the performance of such weather simulations can be improved by allocating subsets of processors to each region of interest instead of the entire processor space

Summary

Introduction

Accurate and timely prediction of catastrophic events such as hurricanes, heat waves, and thunderstorms enables policy makers to take quick preventive actions. Such predictions require high-fidelity weather simulations and simultaneous online visualization to comprehend the simulation output on-the-fly. Ongoing efforts in the climate science and weather community continuously improve the fidelity of weather models by employing higher order numerical methods suitable for solving model equations at high resolution discrete elements. It is necessary to track both depressions to forecast the possibility of a typhoon or heavy rainfall. In such scenarios, multiple simulations need to be spawned within the main parent simulation to track these phenomena. The high resolution simulations are generally executed as subtasks within the coarser-level parent simulation

Methods

Findings

Discussion

Conclusion