Search

The recent rapid advancement of cloud-native networking infrastructure has leveraged the resource virtualization technology of containers to realize diverse microservice-based applications in 5G/6G networks and clouds. Containers drastically enhance the efficiency of computational resource allocation and utilization as compared to the related virtualization technology of Virtual Machines (VMs). The networking environment leveraging both VM and container mixed virtualization technologies makes the most use of them to realize a computational platform whose resources can be dynamically adjusted to a fine granularity. To continuously meet the required levels of quality of services in 5G/6G networks and clouds in that platform, an agile and autonomous data analytics system in the control plane is essential for the accurate prediction of server workloads and dynamic allocation of enough amount of computational resource. In this paper, we introduce a framework, which complies with Recommendation ITU-T Y.3177, for autonomous computational resource control and management. The framework consists of an advanced data analytics system and a resource control system. We propose an architecture for the advanced data analytics system consisting of learning and prediction components. The learning component includes a three-stage intelligent model pipelining with three cascaded machine learning models, nonlinear regression, clustering, and multiple regression. These models determine the fluctuation trends in CPU utilization, classify services with similarities in the trends, and predict the peak CPU utilization of each containerized microservice. We evaluate the proposed models through experiments and numerical analysis. The results prove that the models support agile data analytics, which can complete data processing in the time granularity of seconds and achieve higher prediction accuracy of CPU utilization.

We propose a novel wavelength resource reconfiguration system with a wavelength selective switch (WSS) and our unique burst-mode erbium-doped fiber amplifier (BM-EDFA) for dynamic and agile new-generation optical networks such as an elastic optical network. The system can concurrently establish or remove multiwavelength channels, while preserving the system stability. In a conventional reconfiguration system, the processing time is increased in proportion to the number of the reconfigured wavelength channels because each wavelength channel is sequentially established or removed. On the other hand, in the proposed system, the multiwavelength channels would be simultaneously established or removed. Therefore, in principle, the processing time for reconfiguration of multiwavelength channels can be 1/(the number of the reconfigured channels) compared to the conventional system with sequential processing. In addition, the system can provide a further reduction effect of the processing time since the adjustment of variable optical attenuators of the WSS is unnecessary after establishing or removing multiwavelength channels. We had demonstrated a proof of concept using a reconfiguration system for the first time. The system employing flexible-grid WSSs and a BM-EDFA showed that the total processing time could be more than 100 s shorter in the case of removing four-wavelength channels with the high-quality data transmission.

We have previously presented ChoiceNet, a choicebased architecture for the future Internet. ChoiceNet enables users to choose and pay for advertised competing service component offerings and independently compose the end-to-end services they require. In this paper, we investigate whether we can achieve collaborative optimization (co-optimization) of the network for both the customers and service providers through pre-grooming pricing algorithms within a flexible optical choice-based network. We evaluate the network performance of different path provisioning strategies along with a range of responsive pre-grooming pricing algorithms, and assess whether the network provider and their customers reached co-optimization, by analyzing their impact on the network’s blocking rate of circuit calls, link utilization, and its generated revenue.

We have proposed the concept of an optical packet and circuit integrated network to provide service diversity, energy efficiency and a simplified control mechanism toward new generation networks. In this integrated network, optical data packets and data on lightpaths are transmitted on common physical resources for efficient resource use. In addition, path signaling for lightpath setup and release thorough optical packet switch block is implemented. We set up a primitive optical packet and circuit integrated network including one switching node and a set of packet/path transceiver. We demonstrate 80 (8λ × 10) Gbit/s colored optical packet switching and 8-lightpaths establishment by transferring optical control packets over the optical packet switching.

We newly developed a 3 × 3 integrated optical packet and circuit switch-node. Optical buffers and burst-mode erbium-doped fiber amplifiers with the gain flatness are installed in the 3 × 3 switch-node. The optical buffer can prevent packet collisions and decrease packet loss. We constructed a multi-ring optical packet and circuit integrated network testbed connecting two single-ring networks and a client network by the 3 × 3 switch-node. For the first time, we demonstrated 244 km fiber transmission and 5-node hopping of multiplexed 14-wavelength 10 Gbps optical paths and 100 Gbps optical packets encapsulating 10 Gigabit Ethernet frames on the testbed. Error-free (frame error rate1 × 10(-4)) operation was achieved with optical packets of various packet lengths. In addition, successful avoidance of packet collisions by optical buffers was confirmed.

We developed novel integrated optical packet and circuit switch-node equipment. Compared with our previous equipment, a polarization-independent 4 × 4 semiconductor optical amplifier switch subsystem, gain-controlled optical amplifiers, and one 100 Gbps optical packet transponder and seven 10 Gbps optical path transponders with 10 Gigabit Ethernet (10GbE) client-interfaces were newly installed in the present system. The switch and amplifiers can provide more stable operation without equipment adjustments for the frequent polarization-rotations and dynamic packet-rate changes of optical packets. We constructed an optical packet and circuit integrated ring network testbed consisting of two switch nodes for accelerating network development, and we demonstrated 66 km fiber transmission and switching operation of multiplexed 14-wavelength 10 Gbps optical paths and 100 Gbps optical packets encapsulating 10GbE frames. Error-free (frame error rate < 1×10(-4)) operation was achieved with optical packets of various packet lengths and packet rates, and stable operation of the network testbed was confirmed. In addition, 4K uncompressed video streaming over OPS links was successfully demonstrated.

We present results from the first demonstration of a fully integrated SDN-controlled bandwidth-flexible and programmable SDM optical network utilizing sliceable self-homodyne spatial superchannels to support dynamic bandwidth and QoT provisioning, infrastructure slicing and isolation. Results show that SDN is a suitable control plane solution for the high-capacity flexible SDM network. It is able to provision end-to-end bandwidth and QoT requests according to user requirements, considering the unique characteristics of the underlying SDM infrastructure.

Optical packet and circuit integrated (OPCI) networks provide both optical packet switching (OPS) and optical circuit switching (OCS) links on the same physical infrastructure using a wavelength multiplexing technique in order to deal with best-effort services and quality-guaranteed services. To immediately respond to changes in user demand for OPS and OCS links, OPCI networks should dynamically adjust the amount of wavelength resources for each link. We propose a resource-adjustable hybrid optical packet/circuit switch and transponder. We also verify that distributed control of resource adjustments can be applied to the OPCI ring network testbed we developed. In cooperation with the resource adjustment mechanism and the hybrid switch and transponder, we demonstrate that automatically allocating a shared resource and moving the wavelength resource boundary between OPS and OCS links can be successfully executed, depending on the number of optical paths in use.

We present the first elastic, space division multiplexing, and multi-granular network based on two 7-core MCF links and four programmable optical nodes able to switch traffic utilising the space, frequency and time dimensions with over 6000-fold bandwidth granularity. Results show good end-to-end performance on all channels with power penalties between 0.75 dB and 3.7 dB.

We show on-demand multi-wavelength spectrum and space defragmentation in an SDM and elastic network with four programmable nodes and two multi-core fiber links. The combined approach is shown to reduce blocking and hardware requirements in small nodes.