1. Introduction: With the advancement of computing and communication technologies, data centers support the characteristics of large scalability and high-performance data-intensive supercomputing capacity. Furthermore, next-generation data centers must manage and process exponentially growing volumes of data, which often arrive in time-sensitive streams from arrays of sensors and instruments or as simulation results [1]. Therefore, these urgent flows require real-time response from data centers, which becomes a major challenge for the next generation data center network. For example, healthcare imaging systems illustrate the need for both high data rates and real-time cataloging. Furthermore, high energy physics experiments, which generate high speeds and huge volumes of data that need to be processed and stored in real time [2]. In recent decades, much research effort has been put into data center network architecture to meet the growing data-intensive computing. However, these works only focus on scalability, energy efficiency, etc. [3, 11]. What surprises us is that a temporally predictable network for real-time data-centric computing has not yet been carefully studied. Furthermore, the current TCP/IP-based network architecture is unable to provide temporal predictability characteristics to data-centric processing. TCP/IP and CSMA/CD are the two factors that most influence timing control [4], because TCP/IP is a proven network and CSMA/CD uses an exponential backoff mechanism to avoid collisions. Therefore, it is important to propose a time-predictable network for data-centric processing to support accurate estimation of communication latency.2. Research proposal: To address this… half of the article… et al., Reliable Real-Time Communication in CAN Networks, in the IEEE Transactions on Computers, December 2003.[8]. T. Pop, et al., Timing Analysis of the FlexRay Communication Protocol, in the proceedings of the 18th Euro-micro Real-Time Systems, 2006.[9]. W. Elmenreich, et al., A Comparison of Field bus Protocols: LIN 1.3, LIN 2.0, and TTP/A, in Proceedings of the 10th IEEE Emerging Technologies and Factory Automation, 2005.[10]. R. Kouzes, The changing paradigm of data-intensive computing, in Journal of Computers, vol. 42, 2009.[11]. Turning Challenges into Opportunities in Data Centers, Intel Inc., 2007.[12]. T. Pop, et al., Temporal analysis of the FlexRay communication protocol, in Proceedings of the 18th Euromicro Conference on Real-Time Systems, 2006.[13]. FlexRay Communications System Protocol Specification, Rev. A, Version 2.1, 2005.