Haoran Chen

Re-Provisioning of Advance Reservation Applications in Elastic Optical Networks

Driven by the development of cloud computing and mobile Internet, network-based applications have become diverse. Besides the usual immediate reservation (IR) applications, which need to be served immediately, advance reservation (AR) applications have been introduced to support initial-delay-tolerance services, such as grid computing, virtual machine backup, and so on. The provisioning mechanism of AR applications is different from that of IR applications, because AR requests don’t require to be served immediately, and the network operator must decide the exact serving time in the AR service provisioning process. Based on the scheduling features of AR requests, this paper formulates a resource model and sets up a control framework for elastic optical networks to support AR services. We propose an optimization algorithm with three reprovisioning policies to dynamically reprovision the AR applications, which have already been scheduled. Both the proposed control framework and algorithm are developed on the basis of an open network operating system, which is an open-source software-defined networking controller. Also, the performance of the proposed algorithm is evaluated via software simulation. Demonstration results show the benefits of reprovisioning, and simulation results show that the reprovisioning optimization algorithm can improve the efficiency of network spectrum resource.

Crosstalk-aware RCSA for spatial division multiplexing enabled elastic optical networks with multi-core fibers

In this Letter, we propose two crosstalk-aware routing, core, and spectrum assignment (CA-RCSA) algorithms for spatial division multiplexing enabled elastic optical networks (SDM-EONs) with multi-core fibers. First, the RCSA problem is modeled, and then a metric, i.e., CA spectrum compactness (CASC), is designed to measure the spectrum status in SDM-EONs. Based on CASC, we propose two CA-RCSA algorithms, the first-fit (FF) CASC algorithm and the random-fit (RF) CASC algorithm. Simulation results show that our proposed algorithms can achieve better performance than the baseline algorithm in terms of blocking probability and spectrum utilization, with FF-CASC providing the best performance.

Cross-stratum resource reservation (CSRR) algorithm for deadline-driven applications in datacenter networks

Nowadays, datacenters interconnected with optical networks have become the fundamental infrastructure to accommodate high-performance datacenter applications. Lacking of layer interaction between datacenters and networks during service provisioning, many end user applications cannot efficiently utilize the network capabilities, nor can they achieve the desired quality of service objectives. In response to these challenges, cross-stratum optimization has been studied to optimize the computing and network resources utilization from a united view. Meanwhile, the type of network applications becomes diverse. Compared to traditional immediate reservation (IR) request, a new type of request called advance reservation (AR) has recently been gaining attention for optical networks. IR’s start time of data transmission is assumed to be immediate, while AR request typically specifies the earliest time or the deadline of data transmission. According to the time features of AR request, AR applications should be scheduled in time dimension. Thus, it is both important and challenging to reserve computing and network resources in efficient manners. In this study, a cross-stratum resource model considering time dimension is set up, and a cross-stratum resource reservation (CSRR) algorithm is proposed to schedule AR applications and to reserve cross-stratum resource. Simulation results show that CSRR can reduce the failure rates of AR applications and improve the resource consumption ratio significantly.

Demonstration of multi-domain spectrum defragmentation with minimum controller-participation degree in elastic optical networks

Spectrum fragmentation is a critical issue in elastic optical networks, which leads to inefficient spectrum utilization and high blocking probability. Spectrum defragmentation is considered to conduct reconfiguration of the deployed lightpaths according to different optimization strategies. Compared with the single-domain scenario, spectrum defragmentation in the multi-domain becomes much more complex and challenging due to the necessity of multi-domain cooperation. Based on the hierarchical control architecture of software-defined networking (SDN), this paper proposes a multi-domain defragmentation (MD-DF) scheme with minimum controller-participation degree (CPD). The concept of CPD is introduced to quantize the times that local controllers are involved in multi-domain defragmentation. Based on whether the potential blocked lightpath goes through inter-domain links, the MD-DF scheme includes intra-domain DF and inter-domain DF. The proposed scheme is implemented in the SDN control plane, and experimental results validate its overall feasibility. Meanwhile, the performance of the proposed scheme is evaluated in simulation with three different ratios of intradomain lightpaths. Simulation results indicate that the MD-DF scheme can optimize the utilization of the spectrum resource and achieve a better blocking probability. In addition, this paper also compares intra-domain DF with interdomain DF in terms of operation latency and execution times.

Static Routing and Spectrum Assignment for Deadline-Driven Bulk-Data Transfer in Elastic Optical Networks

Inter-datacenter services such as data replication and virtual machine migration contribute significantly to the underlying network traffic. These bulk-data transfer services can tolerate certain delay if the completion time is guaranteed with a pre-defined deadline. In this paper, we propose joint frequency- and time-domain optimization algorithms of static routing and spectrum assignment (RSA) for these deadline-driven bulk-data transfer requests in elastic optical networks. We formulate the problem as an integer linear program (ILP) and then propose six scheduling schemes, which combine three request ordering strategies with two RSA algorithms. Simulation results show that compared with ILP, our schemes can achieve the same performance in terms of spectrum utilization with efficient time complexity. Furthermore, based on the numerical results and computational efficiency, we provide a guideline on choosing scheduling schemes.

First Field Trial of Virtual Network Operator Oriented Network on Demand (NoD) Service Provisioning over Software Defined Multi-Domain Optical Networks with Multi-Vendor OTN Equipment

A novel NoD service provisioning solution is proposed to satisfy the private network need of virtual network operators. The solution is first verified in the real software defined multi-domain optical networks with multi-vendor OTN equipment.