Factory technologies have evolved to incorporate a great deal of manufacturing flexibility. Programmable automation in the form of CNC and PLCs along with hardware innovations (quick-change tooling, for example) and various operator assist technologies enable a high level of shop-floor flexibility. Possibly, the most inflexible part of a factory is the manufacturing information system. Customized for manufacturers by system integrators, these systems are often large monolithic systems assembled around an ERP/MRP framework or a precariously integrated set of decision-support software tools with a patchwork of communications enabling information flow between them. On the other hand, cloud-based information service platforms such as those encountered in social networks and service brokers have seen rapid and multiple cycles of evolution resulting in a meteoric rise in their ability to handle increasingly large data scales and rates, while still maintaining their elasticity and flexibility. This rapid evolution of cloud-based information services has ignited a new era in the manufacturing industry as evidenced by emerging manufacturing cyberphysical system technologies such as the Industrial Internet of Things (IIoT), and Cloud Manufacturing (CM). These technologies are part of the broader context of what is thought to be the unfolding fourth industrial revolution (Industry 4.0 or Digital Manufacturing). This revolution places at its core, connectivity, information, and machine-based intelligence to create a new paradigm for manufacturing that is highly flexible, scalable, responsive, and intelligent. This paper describes how we leveraged the newest advances in CPS, IIoT, CM, and distributed systems to create a flexible manufacturing information system infrastructure that separates information collection and distribution for decision-making functions. The first part of the paper introduces the architecture for a novel full-stack manufacturing infrastructure that is envisioned to facilitate and track the interaction between a manufacturing job, physical resources, and the software services (or apps) around them. We call this platform the Operating System for Cyber-physical Manufacturing (OSCM). In the second part of the paper, we introduce an event-based architecture for OSCM so that resource or transaction related events/data can be flexibly distributed to different decisionmaking/ manufacturing software tools through an event/message exchange/bus. Further, such an architecture allows modularization and incremental development of different manufacturing software tools and services as new needs are identified.