For nearly two decades, the Controller Area Network (CAN) protocol has been used in the development of distributed embedded systems. Studies previously carried out have illustrated how various “Shared-Clock” (S-C) algorithms can be used with CAN-based microcontrollers to implement time-triggered network architectures. In these previous studies, it has been assumed that the underlying network topology is a bus; in the present paper, we explore some implications of working with a fault-tolerant star-based network topology and a modified S-C algorithm. A representative case study is employed and quantitative comparisons are made between equivalent bus- and star-based systems.