Pengxing Guo

Reliable routing in 3D optical Network-on-Chip based on fault node reuse

Recently, the reliable routing has gradually become the main focus in the on-chip communication field, and some solutions have been applied in Network-on-Chips (NoCs). But with the deepening research on Optical NoCs (ONoCs), the superiority of ONoCs is becoming increasingly obvious. As the most efficient on-chip communication way, ONoC also presents some fault situations due to manufacturing defects. To ensure the operating reliability of ONoCs, a novel adaptive reliable routing algorithm NRFTA (Node Reuse Fault Tolerant Algorithm) is designed by us. After extending the internal structure of Optical Routers (ORs), we utilize NRFTA to tolerate arbitrary number/distribution of the faults from microring resonators. NRFTA can find the restore path within each disabled OR, i.e., fault-node reuse, instead of rerouting packets along the restore path detouring around disabled ORs. Thus, the best macroscopic path along with the least ORs is guaranteed by NRFTA. Simulation results demonstrate that NRFTA performs better than benchmarks, in terms of transmission latency and signal-to-noise ratio.

Designs of 3D mesh and torus optical Network-on-Chips: Topology, optical router and routing module

As a nanometer-level interconnection, the Optical Network-on-Chip (ONoC) was proposed since it was typically characterized by low latency, high bandwidth and power efficiency. Compared with a 2-Dimensional (2D) design, the 3D integration has the higher packing density and the shorter wire length. Therefore, the 3D ONoC will have the great potential in the future. In this paper, we first discuss the existing ONoC researches, and then design mesh and torus ONoCs from the perspectives of topology, router, and routing module, with the help of 3D integration. A simulation platform is established by using OPNET to compare the performance of 2D and 3D ONoCs in terms of average delay and packet loss rate. The performance comparison between 3D mesh and 3D torus ONoCs is also conducted. The simulation results demonstrate that 3D integration has the advantage of reducing average delay and packet loss rate, and 3D torus ONoC has the better performance compared with 3D mesh solution. Finally, we summarize some future challenges with possible solutions, including microcosmic routing inside optical routers and highly-efficient traffic grooming.

Designs of low insertion loss optical router and reliable routing for 3D optical network-on-chip

Three-Dimensional Optical Network-on-Chip (3D ONoC) has recently emerged as a high-performance on-chip communication solution; however, owing to the intrinsic characteristics of photonic devices in existing 3D ONoC, the insertion loss caused by undesirable coupling among optical signals degrades network performance. Considering manufacturing defects and unpredictable noise sources that cause the failure of Optical Routers (ORs) in 3D ONoC, previous work simply abandoned the disabled OR when computing the restore path. In this paper, we propose a new OR structure that reduces insertion loss, and we present the design of a novel adaptive routing algorithm, FTRA-BL, based on the new OR structure with bidirectional waveguides, without abandoning any disabled ORs. Our FTRA-BL selects the normal waveguide in the disabled OR as the backup link so that the best macroscopic restore path can be guaranteed. Simulation results show that our method performs better than previous work in improving transmission reliability and latency.摘要创新点1提出一种新的适合三维 Torus 拓扑的光路由器结构, 具有低插入损耗和高可扩展能力。2.利用几何分析方法对网络中使用的光开关进行建模, 并利用传输矩阵分析光开关插入损耗。3.提出适合光片上网络的可靠性路由策略, 并将故障粒度降低至光路由器内部, 提升了网络抗毁性能。4.首次将单个波导双向传输策略引入到光片上网络中, 并提出双向链路的切换/控制机制。

Experimental demonstration of an intelligent control plane with proactive spectrum defragmentation in SD-EONs

The cooperation of software-defined networking and flexible grid optical transport technology allows operators to elastically control the network using software running on a network operating system within a centralized way. However, existing approaches dealing with spectrum fragmentation are mostly the reactive strategy, which reconfigures network resources to overcome spectrum fragmentation when the controller detects the fragmentation. In this paper, we focus on how to improve the control plane intelligence of software-defined elastic optical networks (SD-EONs) by using a proactive strategy. More specifically, we design a novel routing, modulation level and spectrum allocation algorithm (RMLSA) based on spectral efficiency and connectivity (SEC) i.e., SEC-RMLSA, in order to improve the utilization efficiency of network resources. Meanwhile, we develop a routing application and an extended OpenFlow protocol to achieve a seamless operation between the controller and the optical data plane. Moreover, all the proposed methodologies are implemented and demonstrated in an SD-EON testbed that has both OpenFlow-based control plane and data plane. Finally, the proposed framework, experimental demonstration, and numerical evaluation are reported for different optical flows. The results show the systems overall feasibility and efficiency.

A novel IP-core mapping algorithm in reliable 3D optical network-on-chips

Abstract The Optical Network-on-Chip (ONoC) is considered as a promising way to achieve high performance of multiprocessor systems, and it will be a 3-Dimensional (3D) architecture organized by a certain topology where optical routers are optically interconnected with each other. For the design of 3D ONoCs, the highly reliable IP-core mapping is a key problem of properly assigning IP cores onto optical routers for a given communication task, and it has two main challenges: reliability estimation and mapping scheme. As for reliability estimation, crosstalk noise and thermal sensitivity which severely influence Signal-Noise-Ratio (SNR) should be measured. In addition, although standard genetic algorithms have been widely utilized to solve the optimal mapping solution due to the superiority of simple process, there are some deficiencies such as premature convergence and inferior local searching. In this paper, the impact factors of ONoC reliability are measured by SNR and thermal models, and we also design a novel IP-core mapping algorithm called as CGSA (Cataclysm Genetic-based Simulated Annealing) based on proposed models. In CGSA, we integrate genetic with an improved simulated annealing algorithm assorted with cataclysm strategies, in order to speed up the searching process. Furthermore, to enhance the network reliability, CGSA is bound with the topology selection, i.e., CGSA generates the optimal mapping solution with the best matched 3D ONoC topology. Simulation results show that CGSA is effective on achieving the higher reliability than benchmarks.

Joint Optimization of Latency Monitoring and Traffic Scheduling in Software Defined Heterogeneous Networks

Since the current Internet is only able to provide best-effort services, the quality of service (QoS) for many emerging businesses cannot be well guaranteed. Meanwhile, due to the privatization of networks management, multi-vendor heterogeneous networks are difficult to provide end-to-end QoS assurance on demand. Therefore, heterogeneous devices from different vendors also bring new challenges to the flexible control of network equipment. Software defined network (SDN) is an emerging paradigm which separates the network’s control logic from the underlying routers and switches. In this paper, we design a monitoring loop of link latency by using both LLDP and Echo probing modules. Then, a dynamic routing algorithm is proposed to select optimized transmission path based on the information of link latency. In addition, we develop a routing application assorted with the monitoring mechanism by extending the RYU controller. We implement our solution in a semi-practical SDN testbed. Finally, the overall feasibility and efficiency of the proposed solution are experimentally verified and evaluated.

Maintenance and bandwidth scheduling for cloud rendering in optical data center networks

Animation rendering aims to transform a digital design into a vivid cartoon. To achieve a high-quality animation in the short term, the cloud rendering was presented through consolidating Data Center (DC) resources. Since DCs are far away from each other due to the space limitation, the cloud rendering will run over the Optical Data Center Network (ODCN) where DCs are interconnected by high-speed optical links. This high-speed interconnection well supports the fast data migration for the DC maintenance, which is a preventive measure for safeguarding the uptime of cloud rendering. In this paper, to achieve the least batch maintenance scheduling, multiple DCs are maintained simultaneously through one batch, which makes Rendering Processes (RPs) migrated among DCs via high-speed optical links, and brings the extra costs such as the link-bandwidth consumption. Thus, after performing the least batch maintenance using the heuristic, we determine the bandwidth provisioning for the optical links along a migration trajectory. Simulation results demonstrate the effectiveness of our method.

Networking Big Data: Definition, Key Technologies and Challenging Issues of Transmission

The big data has been touted as the new oil, which is expected to transform our society. Specially, the data source from the networking domain (networking big data) has higher volume, velocity, and variety compared with others. Thus in this article, we make a short survey on existing works investigating key technologies of networking big data, and propose challenging issues of transmission that is the most important stage for networking big data.