Distributed Memory Parallel Implementation of Agent-Based Economic Models

Lalith, M., Gill, A., Poledna, S., Hori, M., Hikaru, I., Tomoyuki, N., Koyo, T., & Ichimura, T. (2019). Distributed Memory Parallel Implementation of Agent-Based Economic Models. In: Lecture Notes in Computer Science. pp. 419-433 Faro, Portugal: Springer. 10.1007/978-3-030-22741-8_30.

Full text not available from this repository.

Abstract

We present a Distributed Memory Parallel (DMP) implementation of agent-based economic models, which facilitates large-scale simulations with millions of agents. A major obstacle in scalable DMP implementation is to distribute a balanced workload among MPI processes, while making all the topological graphs, over which the agents interact, available at a minimum communication cost. We addressed this problem by partitioning a representative employer-employee interaction graph, and all the other interaction graphs are made available at negligible communication costs by mimicking the organizations of the real-world economic entities. Cache-friendly and low-memory intensive algorithms and data structures are proposed to improve runtime and scalability, and the effectiveness of each is demonstrated. The current implementation is capable of simulating 1:1 scale models of medium-sized countries.

Item Type:	Book Section
Uncontrolled Keywords:	Agent-based economic models; Large-scale simulations; MPI
Research Programs:	Advanced Systems Analysis (ASA) Risk & Resilience (RISK)
Depositing User:	Luke Kirwan
Date Deposited:	03 Jul 2019 06:12
Last Modified:	27 Aug 2021 17:31
URI:	https://pure.iiasa.ac.at/15977

Actions (login required)

View Item