A Distributed Architecture for Simulation Environments Based on Game Engine Systems - Visual Computing: Scientific Visualization and Imaging Systems - page 47

Fig. 2.7 Data parallel model

task ordering. In each thread, the tasks for rendering and update are divided taking
into consideration their dependency.
The Data Parallel Model [ 12 ] uses a different paradigm where data are grouped
in parallel sections of the application where they are processed. So, instead of
using a main loop with concurrent parts that process all data, the Data Parallel
Model proposes to use separate threads for sets of data (like game objects). This
way, the objects run their own tasks (like AI and animation) in parallel. Figure 2.7
depicts this approach.