Coordinated process scheduling algorithms for coupled earth system models

Abstract

It is becoming increasingly significant for humans to predict and understand future climate changes using coupled climate system models. Although the performance and scalability of individual physical components have improved over the past few years, coupled climate systems still suffer from low efficiency. This paper focuses on the process scheduling problem for the widely applied coupled earth system model (CESM). The proposed resource allocation strategies allow components to execute on a compromised suboptimal setup and still maintain approximately the best parallel speedup. With this flexible resource allocation strategy, we further propose a coordinated process scheduling algorithm (CPSA). More notably, we propose an upgraded version called CPSA-B, which makes efficient resource sharing configurations, including resource allocation and process layout of components. We integrate CPSA and CPSA-B as pre-arrangement tools into the CESM program and deploy them on the Huawei Kunpeng platform. The speedup curves of the CESM components are prepared in advance, based on sampling tests. Experimental data show that CPSA-B reduces up to 58% of the execution time compared with the CESM default strategy. The algorithm has low complexity and can efficiently find solutions for large input sizes. K E Y W O R D S coupled climate system model, process scheduling, resource allocation, strip packing INTRODUCTIONNowadays, changes of climate have become a critical subject in that they have potentially devastating impact on both our economy and society.Analysis and prediction of future global or regional climate are undoubtedly vital. 1,2 However, climate simulation is a very challenging topic because it involves a large number of resource-consuming physical procedures. 3 In recent years, state-of-the-art climate system models have been developedto integrate observation data with simulation programs of the Earth's climate system, including the atmosphere, ocean, and land surface. 4Coupled climate system models are usually complex and extremely resource-consuming. The National Center for Atmospheric Research reported that an improvement in each percent of the climate simulations in time-to-solution performance frees up to the equivalent of $250,000 in computing resources. 5 Besides, with increasing demands of a larger solution space and extension of simulation grid, these climate models are becoming even more resource consuming.One of these coupled climate systems, the community earth system model (CESM) is a widely applied global climate model. It is a fully coupled system of multigeophysical components. The cyber-physics procedures include the atmosphere component (ATM), 6 ocean component (OCN), 7sea ice component (ICE), 8 and land surface component (LND). 9 The CESM framework executes all these components in an iteration, calculating the results for one simulation day. At the beginning of each iteration, these components periodically exchange model boundaries with a coupler program (CPL). 10