Accelerating parallel agent-based epidemiological simulations

Abstract

Background: Simulations play a central role in epidemiological analysis and design of prophylactic measures. Spatially explicit, agent-based models provide temporo-geospatial information that cannot be obtained from traditional equation-based and individual-based epidemic models. Since, simulation of large agent-based models is time consuming, optimistically synchronized parallel simulation holds considerable promise to significantly decrease simulation execution times.Problem: Realizing efficient and scalable optimistic parallel simulations on modern distributed memory supercomputers is a challenge due to the spatially-explicit nature of agent-based models. Specifically, conceptual movement of agents results in large number of inter-process messages which significantly increase synchronization overheads and degrades overall performance.Proposed solution: To reduce inter-process messages, this paper proposes and experimentally evaluates two approaches involving single and multiple active-proxy agents. The Single Active Proxy (SAP) approach essentially accomplishes logical process migration (without any support from underlying simulation kernel) reflecting conceptual movement of the agents. The Multiple Active Proxy (MAP) approach improves upon SAP by utilizing multiple agents at boundaries between processes to further reduce interprocess messages thereby improving scalability and performance. The experiments conducted using a range of models indicate that SAP provides 200% improvement over the base case and MAP provides 15% to 25% improvement over SAP depending on the model.

Integrated QoS-aware Resource Provisioning for Parallel and Distributed Applications

“…Therefore, they are widely used in various areas, e.g., city transportation [16] and disease propagation [17]. Since a large scale agent-based simulation is time consuming, it is usually partitioned into a number parallel components (federates) which are managed by a parallel simulation framework [18].…”

Optimize Data-Driven Multi-Agent Simulation for COVID-19 Transmission

“…However, computation performance is a significant aspect for large-scale simulations where obstructions from heavy calculation, intensive memory access and communication are inevitable. In general, approaches to improve MAS performance include parallel computing [22,23,24] and distributed computing [25,26,27] on both CPU and GPU platforms. Well known of parallel and distributed platforms contain Mason [28], Gama [29] and Simphony [30].…”

Load-Aware Dynamic Time Synchronization in Parallel Discrete Event Simulation

“…Parallel DES (PDES) refers to the execution of DES on parallel computers and promises fast, scalable and high fidelity simulations of large-scale complex systems. For instance, PDES tools and libraries are employed to analyze the spread characteristics of diseases [39], to design transportation infrastructures [1] and telecommunication systems [21], to simulate hardware description languages [29] and for performance prediction on novel computing platforms [7].…”