Resource utilization of modern data centers is significantly limited by the mismatch between the diversity of the resources required by running applications and the fixed amount of hardwired resources (e.g., number of central processing unit CPU cores, size of memory) in the server blades. In this regard, the concept of function disaggregation is introduced, where the integrated server blades containing all types of resources are replaced by the resource blades including only one specific function. Therefore, disaggregated data centers can offer high flexibility for resource allocation and hence their resource utilization can be largely improved. In addition, introducing function disaggregation simplifies the system upgrade, allowing for a quick adoption of new generation components in data centers. However, the communication between different resources faces severe problems in terms of latency and transmission bandwidth required. In particular,the CPU-memory interconnects in fully disaggregated data centers require ultra-low latency and ultra-high transmission bandwidth in order to prevent performance degradation for running applications. Optical fiber communication is a promising technique to offer high capacity and low latency, but it is still very challenging for the state-of-the-art optical transmission technologies to meet the requirements of the fully disaggregated data centers. In this paper, different levels of function disaggregation are investigated. For the fully disaggregated data centers, two architectural options are presented, where optical interconnects are necessary for CPU-memory communications. We review the state-of-the-art optical transmission technologies and carry out performance assessment when employing them to support function disaggregation in data centers. The results reveal that function disaggregation does improve the efficiency of resource usage in the data centers, although the bandwidth provided by the state-of-the-art optical transmission technologies is not always sufficient for the fully disaggregated data centers. It calls for research in optical transmission to fully utilize the advantages of function disaggregation in data centers.

This paper reports the demonstration of high-speed PAM-4 transmission using a 1.5-mu m single-mode vertical cavity surface emitting laser (SM-VCSEL) over multicore fiber with 7 cores over different distances. We have successfully generated up to 70 Gbaud 4-level pulse amplitude modulation (PAM-4) signals with a VCSEL in optical back-to-back, and transmitted 50 Gbaud PAM-4 signals over both 1-km dispersion-uncompensated and 10-km dispersion-compensated in each core, enabling a total data throughput of 700 Gbps over the 7-core fiber. Moreover, 56 Gbaud PAM-4 over 1-km have also been shown, whereby unfortunately not all cores provide the required 3.8 x 10(-3) bit error rate (BER) for the 7% overhead-hard decision forward error correction (7% OH-HDFEC). The limited bandwidth of the VCSEL and the adverse chromatic dispersion of the fiber are suppressed with preequalization based on accurate end-to-end channel characterizations. With a digital postequalization, BER performance below the 7% OH-HDFEC limit is achieved over all cores. The demonstrated results show a great potential to realize high-capacity and compact short-reach optical interconnects for data centers.

With the approach of the 5G era, stringent requirements are imposed on the data transport solutions, including both of the supported transmission reach and the capacity. Radio-over-fiber technologies are considered to be promising candidates to cope with both aspects, owing to the low-loss and broad-bandwidth nature of the optical fibers. Meanwhile with such optical transport solutions, signals can be collected from the distributed remote radio sites and processed in a centralized manner. In this report, we target on the digital radio-over-fiber systems, and discuss about several key technologies, focusing on the aspects of coding and transmission, which could potentially enable terabit-scale data transport.

Quantum key distribution (QKD) networks are promising to progress towards widespread practical deployment over existing fiber infrastructures in the near future. Given the high cost and difficulty of deploying QKD networks, multi-tenancy becomes promising to improve cost efficiency for future QKD networks. In a multi-tenant QKD network, multiple QKD tenants can sham the same QKD network infrastructure to obtain secret keys for securing their data transfer. Since the secret-key resources are finite and precious in QKD networks, how to achieve efficient multi-tenant secret-key assignment (MTKA) to satisfy the secret-key demands of multiple QKD tenants over QKD networks becomes a significant problem. In this regard, this study addresses the MTKA problem over QKD networks. A new multi-tenant QKD network architecture is proposed based on software defined networking (SDN) and quantum key pool (QKP) techniques. A secret-key rate sharing scheme is presented and a heuristic algorithm is designed to implement efficient MTKA over QKD networks. A new performance metric, namely matching degree (MD) that reflects the balance between QKD network secret-key resources and QKD tenant requests, is defined and evaluated. Simulation studies indicate that high QKD tenant requests accommodation and efficient secret-key resource usage can be achieved via maximizing the value of MD. 

We experimentally demonstrate the generation of 107-Gbit/s net-rate optical discrete multitone signal using a 1.5-mu m single-mode vertical cavity surface emitting laser (VCSEL) with modulation bandwidth of 22 GHz. Utilizing a nonlinearity-tolerant channel equalization algorithm for digital signal processing, total net-rates of 726.6-Gbit/s over 2.5-km dispersion-uncompensated 7-core fiber and 533.1-Gbit/s over 10-km dispersion-compensated 7-core fiber below 7% overhead hard-decision forward error correction limit have been experimentally achieved with a 1.5-mu m VCSEL-based intensity-modulation direct-detection system. The features of the 1.5-mu m single-mode VCSEL, 2.5-km/10km multi-core fibers, and fan-in/fan-out modules are presented. Besides, the Volterra series-based nonlinearity-tolerant channel equalization algorithm, which improves the signal-to-noise ratio with more than 5 dB, is mathematically described and experimentally validated. The results have demonstrated that 1.5-mu m single-mode VCSEL and multi-core-fiber-based transmission can be a promising candidate to solve the capacity challenges in short-reach optical interconnects.

To support massive deployment of broadband radio applications, such as 5G and high-definition videos for terrestrial televisions, large system capacity and high spectrum efficiency are highly demanded in radio over fiber (RoF) systems. In this article, we propose a terabit digital RoF system capable of providing high-speed transmission, where multicore fiber (MCF) is introduced for the access segment between the central unit and remote unit. Two key technologies that greatly enhance system capacity and spectrum efficiency, namely MCF enabled self-homodyne detection and compressed quantization, are demonstrated.

We demonstrate 56 Gbaud/λ PAM4 inter - data center interconnects over 81 km single core single mode fiber and 33.6 km 7-core single mode fiber with continuous-fiber Bragg grating based chromatic dispersion compensators covering C-band.

100 Gbps/λ/core PAM-4 transmission is successfully demonstrated over 1-km and 10- km single mode 7-core fiber links, enabled by directly modulated 1.5-μm single mode VCSEL of 23 GHz modulation bandwidth with pre- and post- digital equalizations.

We transmit 80 Gbaud/λ/core PAM4 signal enabled by 1.55 μm EML over 1 km 7-core fiber. The solution achieves single-wavelength and single-fiber 1.04 Tbit/s post-FEC transmission enhancing bandwidth-density for short-reach optical interconnects.

A 107Gb/s net-rate DMT optical signal was generated using a single-mode long-wavelength VCSEL with a modulation bandwidth of 23GHz. We experimentally demonstrated a total net-rate up to 726.7Gb/s at 1.5 mu m over 2.5km 7-core dispersion-uncompensated MCF.