Distributed systems based on “time triggered” (TT) “Shared-Clock” (SC) architectures are the main concern in the research described in this thesis. Such architectures are often employed in safety-critical embedded systems because–if implemented correctly–they can provide a foundation for designs which have very predictable patterns of behaviour. Previous research in this area has explored the development of both single and multi-processor TT designs. In the case of multi-processor designs, the focus has been on “Shared-Clock” (SC) architectures.In SC designs, the Controller Area Network (CAN) protocol – introduced by Robert Bosch GmbH in the 1980s – can provide high levels of reliability at low cost. As a consequence, the CAN protocol has become widely used in many sectors (e.g. automotive and automation) and almost all modern microcontroller families now support this protocol. All previous work on SC protocols has considered designs based on a bus topology. The target of this research was to explore other possibilities by developing novel SC protocols based on a novel star topology. The work had two main motivations: (1) to improve the flexibility of such designs significantly, by facilitating the creation of systems with “tick rates” flexibility on each arm of the star; (2) to improve the reliability of designs based on a shared-clock protocol. In this thesis, three “Time-triggered Co-operative, Shared-Clock” (TTC-SC) protocols are introduced: these are referred to as “TTC-SC5” and “TTC-SC6” which were developed previously. As a contribution of this thesis, the culmination of those two previously developed protocols gave rise to our third novel protocol called the “enhanced TTC-SC7” that embodies capabilities of both its predecessors. The TTC-SC5 protocol was previously developed to address the challenges of co-operative scheduling in TTC-SC designs. TTC-SC5 addressed such challenges through a new strategy known as the “Differential Tick Rate” (DTR) mechanism. Also, the TTC-SC5 protocol countered the Single-Point-of-Failure (SPF) hypothesis for the novel star topology described later in this thesis. As CAN-related hardware has an inherent fault-model, addressing such faults is crucial for the normal operation of SC architectures. Building on TTC-SC5, the TTC-SC6 protocol was developed previously to add additional support for fault management in CAN networks based on a star topology. The TTC-SC6 protocol achieved its fault-confinement and fault-tolerance capabilities through a new strategy which was known as the “Port Guardian” (PG) mechanism. In this thesis, it is argued that the amalgamation of our previously developed techniques in TTCSC5 and TTC-SC6 can considerably improve the flexibility as well as reliability of CAN based distributed systems that employ a shared-clock architecture through our novel enhanced TTCSC7 protocol in one suit. A comparative analysis of the software codes used for CANbus and the migrated CANstar based SC architecture is also a part of this thesis. With such a comparison, we intend to show that code wise bus to star migration through our enhanced TTC-SC7 protocol is easily achievable with less software complexity than the former one.