Additive manufacturing of metallic components is an emerging technique that enables the customized production of components with complex shapes, making it highly valuable for aerospace, automotive, and biomedical applications. However, many engineering alloys are not suitable for this process due to hot cracking, which is linked to the microstructure of large columnar grains. A microstructure with fine equiaxed grains has proven effective in avoiding the hot cracking problem. Furthermore, it is crucial to engineer the texture of additively manufactured alloys to control the anisotropy of their properties. Therefore, we are developing a modeling framework to predict microstructure in additive manufactured components, starting from solidification conditions to determine the nucleation and growth of grains and finally the size distribution and texture of grain structure can be calculated. This framework has proven to be computational affordable at meltpool (~0.5 mm) level, using normal PCs. However, experimental works have shown that by designing scanning strategies during additive manufacturing, substantial differences in grain structure at component level (several millimeters or above) can be achieved. To be able to predict microstructure at component scale within reasonable time, using HPCs will be necessary. Therefore, we are proposing to use a combination of various modeling tools at different length scales to simulate grain structure of additive manufactured components for different alloy compositions and scanning strategies during additive manufacturing using HPCs.