Wenda Ni

POXN: A New Passive Optical Cross-Connection Network for Low-Cost Power-Efficient Datacenters

Passive optical devices, characterized by low cost, zero energy consumption, and high reliability, are essential building blocks for todays telecom network infrastructure, permeating from conventional backbone transport networks towards next-generation broadband access networks. Motivated by the striking features of passive optical devices, in this paper, we seek their potential applications in emerging datacenter networks to tackle the scalability challenges arising from cost and power. Specifically, we propose passive optical cross-connection networks (POXNs) that enable cost-saving, power-efficient, and reliable communication within datacenters. To support POXNs in warehouse-scale datacenters, we address physical-layer scalability challenges by using advanced interconnection techniques. Next, we propose a distributed polling protocol to address link-layer issues that arise from the broadcast nature of the medium. The performance of our protocol is studied through analysis and simulation. In particular, we develop an analytical model to compute lower and upper bounds on the expected delay of a packet. Numerical results show that the mean packet delay is equal to the lower bound in one regime, while converges to the upper bound in the complementary regime. Results also show that our protocol can achieve high bandwidth efficiency (no less than 85% in our studied case). Additionally, we demonstrate that our protocol can embrace scheduling algorithms that support fairness and QoS. Last, we sketch the roles POXNs can play in various datacenter network architectures in terms of capital and operational cost reductions.

Provisioning high-availability datacenter networks for full bandwidth communication

One critical challenge in datacenter network design is full bandwidth communication. Recent advances have enabled this communication paradigm based on the notion of Valiant load balancing (VLB). In this paper, we target full bandwidth communication among all servers, for all valid traffic patterns, and under k arbitrary link failures. We focus on two typical datacenter topologies, VL2 and fat-tree, and propose a mechanism to perform VLB on fat-tree. We develop the minimum link capacity required on both topologies, where edge and core links are handled separately. These results can help datacenter providers to provision their networks with guaranteed availability. Based on the results, we evaluate the minimum total link capacity required on each topology and characterize the capacity increase trend with k and with the total number of supported servers. These studies are important for datacenter providers to project their capital expenditures on datacenter design, upgrade, and expansion. Next, we compare the total link capacity between the two topologies. We find that given the same server scale, fat-tree requires less total capacity than does VL2 for small k. For large k, there exists a turning point at which VL2 becomes more capacity-efficient.

Availability Analytical Model for Permanent Dedicated Path Protection in WDM Networks

This paper analyses the connection availability of a protection paradigm, termed permanent dedicated path protection (P-DPP), in WDM mesh networks. An analytical model for calculating the unavailability of P-DPP is introduced. Numerical results on COST239 European network show that the average unavailability of the network is proportional to blocking probability of the network, and fits well with the simulation results.

Availability of survivable Valiant load balancing (VLB) networks over optical networks

Valiant load balancing (VLB) network has been proposed as a capacity-efficient solution to handle highly dynamic traffic in future backbone networks. In this paper, we study the availability of VLB networks that are overlaid over an optical infrastructure. The main challenges in such a context arise from the unique routing and protection scheme that goes beyond the definition of conventional connection-level service availability as well as the logical link failure correlation that prohibits the use of traditional analytical methods. We propose a network-level availability model to compute the probability that a VLB network is congestion-free under all traffic patterns. Numerical results show that with a proper truncation level, our calculation on availability can be accelerated significantly by generating tight lower and upper bounds. Our main finding is that physical link sharing in a two-layer setting degrades the network availability drastically by several orders of magnitude due to the full mesh requirement for VLB networks, and may remove the capacity efficiency advantage of VLB networks.

Analytical models of flow availability in two-layer networks with dedicated path protection

Traffic flows with protection can be disrupted in network failures that are beyond the protection capability. We analyze availability of upper-layer flows in two-layer networks with dedicated path protection at upper or lower layer.

Optimizing the monitoring path design for independent dual failures

This paper proposes a new monitoring path design paradigm for independent dual link failures. Specifically, the new approach exploits the sequential arrival and departure property of independent failure events to uniquely localize failed links. Such property, however, cannot be captured by the existing approach, which is built upon the notion of shared risk link groups. Consequently, we show via solution space comparison that the existing approach can result in overdesign in terms of monitoring resources required. Numerical results further indicate that the new approach outperforms the existing one in terms of monitoring cost and computational efficiency.

Availability analytical model for permanent dedicated path protection in service differentiated WDM networks

This paper analyses the connection availability of a protection paradigm, termed permanent dedicated path protection, in service differentiated WDM mesh networks. An analytical model for calculating the unavailability of P-DPP is introduced.

On capacity provisioning in datacenter networks for full bandwidth communication

Recent advances in datacenter network design have enabled full bandwidth communication based on the notion of Valiant load balancing. In this paper, we study from the link capacity viewpoint how such communication paradigm can be supported in the context of link failures, the study of which is currently absent. In particular, we target full bandwidth communication among all the servers, for all valid traffic patterns, and under k arbitrary link failures. We derive the minimum link capacity required on two typical datacenter topologies-VL2 and fat-tree. Our main finding is that given the same server scale, fat-tree requires less total link capacity than VL2 for 1 ≤ k ≤ n/4, where n denotes the port count of homogeneous switches used in fat-tree. For k > n/4, there exists a turning point from which VL2 becomes more capacity-efficient.

Incremental survivable network design against node failure in SDH/SONET mesh networks

Network survivability is becoming more and more important for the plenty of information each single fiber carries. Extra network resources are needed to increase network survivability level. In this paper, we investigate the problem of how to augment the network topology with adding new links and allocate spare capacity to maximize the service restorability against node failures in SDH/SONET mesh networks. A scheme called maximal node-disjoint backup paths provisioning with topology augmentation is proposed to tackle the problem, and another scheme called globally optimized path provisioning with topology augmentation, which allows adjusting the existing working paths of network flows, is investigated to optimize the augmented network globally. Both schemes are formulated as mixed integer linear programming models. Furthermore, heuristic algorithms are investigated to be implemented in software. Three algorithms, i.e., added links searching method, successive maximal survivable routing method, and random sequence routing convergence method, are designed and compared. Simulation results show the effectiveness of the algorithms.

Integrated design of fault localization and survivable mapping in IP over transparent WDM networks

Lightpaths play multiple critical roles in cross-layer fault management of IP over transparent optical networks. While the roles of lightpaths have been discussed specifically in fault localization as well as fault recovery phases, these existing works only take a sporadic, piecemeal view of these roles by studying each of them individually as a separate lightpath routing problem. In this paper, we instead take an integrated, systematic view by considering these multiple roles jointly as one single lightpath routing problem. In particular, we propose a new design model to fulfill the lightpath layout requirements in both identifying fiber link failures for fault localization and ensuring IP topology connectivity for fault recovery. With a much smaller formulation size, our new model significantly outperforms the existing counterpart in computational efficiency, scalability, and solution quality.