Abstract:The graphic processing unit (GPU) can perform some large-scale simulations in an economical way. However, harnessing the power of a GPU to discrete event simulation (DES) is difficult because of the mismatch between GPU’s synchronous execution mode and DES’s asynchronous time advance mechanism. In this paper, we present a GPU-based simulation kernel (gDES) to support DES and propose three algorithms to support high efficiency. Since both limited parallelism and redundant synchronization affect the performance of DES based on a GPU, we propose a breadth-expansion conservative time window algorithm to increase the degree of parallelism while retaining the number of synchronizations. By using the expansion method, it can import as many as possible ‘safe’ events. The irregular and dynamic requirement for storing the events leads to uneven and sparse memory usage, thereby causing waste of memory and unnecessary overhead. A memory management algorithm is proposed to store events in a balanced and compact way by using a lightweight stochastic method. When events processed by threads in a warp have different types, the performance of gDES decreases rapidly because of branch divergence. An event redistribution algorithm is proposed by reassigning events of the same type to neighboring threads to reduce the probability of branch divergence. We analyze the superiority of the proposed algorithms and gDES with a series of experiments. Compared to a CPU-based simulator on a multicore platform, the gDES can achieve up to 11×, 5×, and 8× speedup in PHOLD, QUEUING NETWORK, and epidemic simulation, respectively.

Research on Performance Based on Time Warp in Parallel Discrete Event Simulation System

Research for Minimum Time Windows Algorithm in Parallel and Distributed Simulation

Time Synchronization in Parallel Simulation Platform

Survey on Synchronization Mechanism in Parallel Discrete Event Simulation

Time Management in Parallel Discrete Event Simulation

A Well-Balanced Time Warp System On Multi-Core Environments

Exploiting Task-Based Parallelism for Parallel Discrete Event Simulation

Research on technologies of Event Graph based Parallel Discrete Event Simulation

Research on Lookahead Problem in Parallel and Distributed Simulation

A GPU-based Discrete Event Simulation Kernel

TIB: A Time Management Algorithm of PDES for Automatically Detecting Concurrency

Design of PDES Assignment Dispatching and Scheduling Tool for Multi- replication

A Global Schedule Mechanism for PDES on Multi-Core Environments

Study of Scalable Sensor Network Simulator and Performance Prediction

Enhanced Null Message Algorithm for PDES with Diverse Event Density.

Derivative estimates parallel simulation algorithm based on performance potentials theory

Performance Optimization for Parallel Systems with Shared DWM Via Retiming, Loop Scheduling, and Data Placement

Intelligent Rollbacks Analysis Algorithm For Optimistic Parallel Discrete Event Simulation

Study of PDES Testing Model

Parallel Discrete Event Simulation Method Based on Event Graph

Study and Implementation of a Hierarchical Time Management Mechanism in Agent-Based Distributed Simulation