In situ visualization is important to meet the needs of scientific visualization for exascale computing in diverse areas of science and engineering. Knowledge discovery from such simulations remains challenging and it is becoming increasingly urgent as the trend toward ultrascale computing with millions of cores continues. Under current trends, most simulated data could be lost if in situ visualization is not deployed. Therefore, traditional numerical simulation data output and post-processing visualization might be unsustainable. In situ visualization in this paradigm will gain unprecedented access to simulation output, potentially being able to process all relevant simulation output at every simulation time step, allowing for very high temporal fidelity compared with traditional post-processing visualization. The solution is to minimize data storage by extracting important features of the data at high temporal resolution using in situ visualization of tight coupling (on the same resource as the simulation), and saving them rather than the raw data. This paper adopts the component-based software engineering mode to realize tight coupling of in situ visualization with a simulation program. It resolves key technological issues of in situ visualization processing, e.g., data access, load balance, and predefined visual analysis patterns. This paper presents preliminary results of in situ visualization of a simulation of large-scale filamentous phenomena in laser plasma under the black cavity physical condition. This simulation is used to study the beam divergence phenomenon after laser filament formation. Moreover, we examine the overheads and effects on code performance associated with inline computations associated with in situ visualization.