In the DC 3.0 architecture white paper, Huawei describes per capita consumption data as a barometer of social development and discusses the series of challenges faced by existing data centers in preparing for future big data processing, for example in real-time PB/s level data processing PB/s, on-demand dynamic resource allocation and adjustments, and optimizing the energy efficiency of large-scale systems. As a solution, Huawei proposes data-centric High Throughput Computer DC 3.0 architecture to build next-generation data centers, via core technologies such as resource pooling, hardware decoupling, optical interconnects and data center-level programming frameworks. Using non-volatile memory and silicon photovoltaic technology to achieve high-bandwidth low-latency network memory as a service and unity in internal and external memory, the DC 3.0 architecture breaks the traditional input/output bottlenecks, while providing more fine-grained resource management and server configuration to resolve fragmentation and improve resource utilization rates. Through this series of technological innovations, Huawei aims to provide an enhanced platform for future large-scale data processing, in order to achieve real-time data processing per second PB level and energy efficiency that is more than 10 times that of existing systems.
Huawei’s DC 3.0 architecture white paper was pre-released at the International Symposium on Computer Architecture (ISCA) held on June 14-18, 2014, at which it received a positive response from the industry and academia.
Huawei's continued investment and innovation in the IT field has seen it evolve from the position as industry “challenger” to that of industry “innovator”. Huawei is committed to continuing its focus on the future development of IT to build core competitiveness, and to provide customers with future-oriented business solutions.