System on Chip (SoC) design in the forthcoming billion-transistor era will involve the integration of numerous heterogeneous semiconductor intellectual property (IP) blocks. Some of the main problems in the ultra deep submicron technologies arise from non-scalable global wire delays, failure to achieve global synchronization and difficulties associated with non-scalable bus-based functional interconnect. These problems can be dealt with by using a structured interconnect template to design future SoCs. Recently, we introduced the butterfly fat-tree as an overall interconnect architecture, where IPs reside at the leaves of the tree and switches at its vertices. Here, we analyze this architecture with a particular focus on achieving overall timing closure. The only global wires in this routing architecture are the inter-switch wires and the delays in these global wires can be predicted at the initial stages of design cycle. Our analysis shows that the inter-switch wire delay in the networked SoC can be always designed to fit within one clock cycle, regardless of the system size. We contrast the analysis for our network with that of a bus-based architecture. For the latter, we illustrate how the interconnect delay and system size are interrelated, thereby limiting the number of IP blocks that can be connected by a bus.