Data Centre Connectivity Is Key to Developing the SIJORI AI Corridor
Connectivity between data centres in Singapore, Johor, and Batam is critical to realising the potential of the SIJORI AI corridor, beyond just building compute capacity.
CommScope highlighted that building data centres in Singapore, Johor, and Batam alone will not create an effective AI corridor unless these facilities are linked by resilient, high-capacity networks that enable seamless interaction across the locations.
Quick Skim
Interconnectivity Supports AI Workload Demands
The SIJORI region, comprising Singapore, Johor in Southern Malaysia, and Batam in Indonesia, presents strategic advantages for Southeast Asia’s digital infrastructure. However, data centres must function within an interconnected ecosystem to fully leverage AI workloads.
AI applications generate diverse network traffic that stresses multiple infrastructure layers, from long-haul backbone to metro networks and data-centre-to-data-centre links. Model distribution, inference fan-out, and synchronisation between GPU clusters require low latency and high throughput connections.
Physical Network Design Needs to Match AI Requirements
While data-centre capacity is often measured in megawatts, CommScope emphasises that connectivity parameters such as physical route diversity, fibre counts, latency, and rapid restoration capabilities are equally essential.
Long-haul routes may involve fibre counts between 144 and 288 per route, metropolitan networks can require up to 864 fibres, and intra-data-centre connections may demand fibre infrastructure exceeding 6,000 fibres. Network resilience requires multiple diverse paths and restoration measured in hours.
Designing resilience into networks from the outset is crucial. Redundant fibre routes following the same physical path offer limited protection, and extensive testing—including OTDR and dispersion analyses—is necessary to ensure performance before deployment.
End-to-End Corridor Connectivity Underpins Economic Value
CommScope stated that AI network traffic transcends organisational boundaries, traversing data halls, metro networks, long-haul routes, and international gateways. Thus, infrastructure planning must consider the entire corridor holistically.
This approach moves beyond individual data-centre projects to focus on linking the SIJORI locations into a consistent, reliable digital infrastructure ecosystem. The capacity to interconnect computing facilities effectively will determine the corridor’s future economic impact.
According to CommScope, the fundamental question for the region’s AI infrastructure development shifts from how much computing power can be built, to how well these computing centres can be connected and operated with resilience and efficiency. This focus on connectivity is critical for Southeast Asia’s evolving AI landscape.


