Sarcouncil Journal of Engineering and Computer Sciences

Abstract: This article examines the conspicuous absence of native IP multicast support in major hyperscale cloud environments, with a particular focus on Google Cloud Platform (GCP). While IP multicast offers an efficient mechanism for one-to-many data distribution in traditional networks, implementing these protocols in modern cloud architectures presents significant challenges. Through analysis of GCP's Andromeda network virtualization stack, the fundamental architectural conflicts between traditional multicast protocols and hyperscale Software-Defined Networks are identified. The inherent requirements of massive scale, multi-tenancy, and network dynamism in cloud environments create technical barriers that lead providers to favor alternative solutions. The article explores several alternative approaches cloud platforms employ to address multicast requirements, including application-level replication, user-managed overlay networks, managed messaging services, content delivery networks, and software-defined multicast implementations. Future directions that might eventually reconcile multicast with cloud architectures are also examined. The absence of native multicast represents a deliberate architectural trade-off prioritizing scalability, security, and operational simplicity over theoretical network efficiency.