> **来源：[研报客](https://pc.yanbaoke.cn)** # Data Centre Futures Summary: Data and Compute ## Core Content This report explores how the evolving nature of data and compute is redefining the design and location of data centres. It emphasizes the need to move beyond viewing data centres as mere assets and instead consider them as critical nodes in urban and global systems, capable of positive social, environmental, and economic impact. The focus is on long-term adaptability, resilience, and alignment with the local context, ensuring data centres can serve as both a "good neighbour" and a "good ancestor." ## Main Themes ### 1. **Data Centres as Urban Infrastructure** - Data centres are essential parts of the urban metabolism, influencing the flow of water, energy, materials, and people. - They are increasingly seen as strategic assets, requiring careful location planning for resilience and autonomy. - The design of data centres must account for their long-term role in shaping sustainable, regenerative urban environments. ### 2. **Adaptive and Context-Specific Design** - Traditional data centres are being replaced by modular, heterogeneous, and reconfigurable infrastructures. - The coexistence of classical, quantum, and biocomputing paradigms is driving the need for specialized zones within a campus. - Location strategies must evolve to match the unique needs of different computing paradigms, such as geologically stable zones for quantum and proximity to helium supply chains. ### 3. **Shift to Edge Computing and Distributed Networks** - AI is evolving from content generation to action execution, leading to a shift from centralized to distributed compute. - "Ladder-shaped" data centre networks are emerging, with local edge nodes, regional hubs, and core sites for training and storage. - These networks improve system resilience by decentralizing critical operations while maintaining efficient large-scale processing. ### 4. **Storage Evolution and Tiered Systems** - Storage is moving from a one-size-fits-all model to optimized tiers based on data sensitivity and access frequency. - New archival media like DNA and glass-based storage offer higher density and lower energy use, though are not yet widely deployed. - Tiered storage systems (hot, warm, cold) are being implemented to reduce costs and environmental impact, with security becoming a key driver. ### 5. **Security and Sovereignty as Key Considerations** - Data sovereignty and legal regulations are shaping where data can be stored and processed. - Security is now a core design factor, with quantum-safe encryption and physical separation becoming essential. - Location decisions must account for regulatory environments, geopolitical tensions, and the need for secure, isolated systems. ### 6. **Future-Ready and Flexible Infrastructure** - The industry is moving toward more flexible and adaptable infrastructure to avoid lock-in and redundancy. - Modular designs allow for easier integration of new technologies without full redesign. - Operators are rethinking how to align compute and storage with local energy, water, and governance realities. ### 7. **Emerging Technologies and Their Impact** - A range of emerging technologies, including AI, quantum computing, biocomputing, and photonic computing, are reshaping the data lifecycle. - These technologies introduce new spatial, operational, and investment requirements, necessitating a more nuanced approach to infrastructure planning. - The report uses a technology radar to map these changes across short, medium, and long-term horizons. ## Key Drivers of Change | Driver | Description | |--------|-------------| | **Compute Shifts Outward** | AI and real-time processing are pushing compute closer to the edge, creating a more distributed network. | | **Storage Shifts to Optimized Tiers** | Storage is becoming more specialized, with different media types supporting varying data access and security needs. | | **Security and Sovereignty** | Legal and geopolitical factors are driving the need for localized, secure data infrastructure. | | **Facilities Shift to Specialized Environments** | Divergent computing demands require data centres to be designed for specific hardware and environmental conditions. | ## Early Signs of the Future - **Barcelona, Spain**: A city-scale edge computing network using 18,000 sensors enables local data processing and reduces reliance on centralized systems. - **Vevey, Switzerland**: Neuroplatform uses lab-grown neural organoids for "wetware computing," demonstrating the potential of biocomputing. - **DNA-Based Storage**: Catalog is developing DNA storage for long-term, high-density data archiving, reducing the need for large physical facilities. ## Recommendations ### ACT NOW - **Location Strategy**: Plan portfolios, not single sites. Define roles for edge, regional, and core data centres based on function and context. - **Adaptability**: Design for flexibility and reconfigurability to accommodate future technological changes. - **Efficiency**: Improve energy and resource efficiency through advanced cooling, modular layouts, and optimized storage. - **Community Engagement**: Align data centre operations with local needs, environmental constraints, and governance structures. - **Resilience and Longevity**: Ensure data centres are resilient and leave a positive legacy, avoiding stranded infrastructure. ## Conclusion The future of data centres lies in their ability to adapt to changing technological, regulatory, and environmental landscapes. By rethinking their role as "good neighbours" and "good ancestors," the industry can move toward more sustainable, equitable, and context-aware digital infrastructure. This report encourages a shift in mindset, from fixed assets to dynamic, flexible systems that support both current and future needs.