Fig. 2
From: Parallelisation strategies for agent based simulation of immune systems
FLAME GPU state diagram for the particle implementation of pairwise cell interaction model. The diagram shows the order of agent functions (black rectangles) and interactions between agents via message lists (coloured rhombuses) through a single iteration of the main simulation loop. Within a layer, indicated by blue dashed boxes, functions may execute concurrently. In the first function layer A agents and B agents both execute their respective move behavioural function, where individuals move using Brownian motion. In the second layer, A agents execute the Output Location function, broadcasting their location within the simulation environment to the A Location message list. This message list is iterated by B agents in the Select + Declare Target function, where B agents select the A agent they wish to interact with and broadcast it into the B Interaction message list. In the fourth layer, A agents iterate the B Interaction message list, deciding which B interaction they will participate in and output a message as confirmation. Finally B agents iterate the confirmation messages to discover if they were successful, and if so they behave appropriately