Xiaoning Zhang

Local Fast Reroute With Flow Aggregation in Software Defined Networks

In this letter, we propose a local fast reroute (LFR) algorithm with flow aggregation in software defined networks (SDN). In LFR, if a link failure is detected, all traffic flows affected by the failure are aggregated to be a new “big” flow. Then, a local reroute path is dynamically deployed by the SDN controller for the aggregated flow. LFR reduces the number of flow operations between the SDN controller and switches. Numerical results show that the LFR enables fast recovery while minimizing the total number of flow entries in SDN.

Performance-aware Energy-efficient Virtual Machine Placement in Cloud data center

This paper studies how to consolidate Virtual Machines (VMs) on physical severs in order to minimize the energy consumption in Cloud data centers. In Cloud data centers, a major energy consumption component is the physical servers, and it is a way to save the energy by consolidating VMs onto fewer VMs and shut down the idle physical servers. However, executing more VMs on fewer physical servers should degrade the VM performance. Accordingly, how to consolidate VMs but guarantee the VM performance is an important issue to study. To this end, we first formulate the VM consolidation problem as a mathematic model and then prove that it is equivalent to a generalized constrained minimum fc-cut problem, which is the hardest NP-hard problem. Due to the problem complexity, an efficient heuristic named PEVMP (Performance-aware Energy-efficient Virtual Machine Placement) is proposed to solve it. In addition, we extend the PEVMP to be an online version to handle service dynamics. Extensive simulation results show that PEVMP can save up to 36.78% energy, and the performance of online algorithm is close to the offline result.

Flow Entry Sharing in Protection Design for Software Defined Networks

In Software Defined Networks (SDN), though we can design protection mechanism to enable fast recovery against a single link failure, it requires proactively installing a large number of flow entries in switches on working paths and backup paths. However, these additional flow entries in a SDN switch may exhaust Ternary Content Addressable Memory (TCAM) which is limited in size since it is expensive and power hungry. Accordingly, it is emergent to design a new protection technology to minimize flow entry occupation. To this end, we leverage flow entry sharing in SDN protection to solve this problem. We first present the problem as an ILP (Integer Linear Programming) model, and then design a greedy based heuristic algorithm named Flow Entry Sharing Protection (FESP). Extensive simulation results show that compared with the previous SDN protection algorithms, FESP reduces 28.31% flow entries and 27.65% link bandwidth in average.

OSNR penalty-free add/drop performance of DSP-enabled ROADMs in coherent systems

The newly proposed digital-signal-processing (DSP)-enabled reconfigurable optical add/drop multiplexer (ROADM) without incorporating optical-electrical-optical conversions, referred to as a soft-ROADM, performs dynamic, flexible, and transparent add/drop operations at wavelength, sub-wavelength, and spectrally overlapped orthogonal sub-band levels. The cost-effective soft-ROADMs offer vital software-defined-networking-based networking functionalities for seamlessly integrating optical and mobile networks to realize elastic cloud access networks. In this paper, detailed theoretical and numerical investigations are undertaken of soft-ROADM add/drop operation performance in coherent systems. It is shown that the soft-ROADM add/drop operation performance is independent of both signal spectral location and fiber transmission distance and also shows excellent robustness to variations under system operating conditions. More importantly, under practically achievable system conditions, the soft-ROADM add/drop operations introduce negligible optical signal-to-noise ratio (OSNR) penalties. Moreover, numerical explorations of the impacts of key physical aspects associated with both the soft-ROADMs and coherent systems are also conducted, based on which optimum soft-ROADM design parameters are identified.

Congestion-aware local reroute for fast failure recovery in software-defined networks

Although a restoration approach derives a reroute path when failure occurs and greatly reduces forwarding rules in switches compared with a protection approach, software-defined networks (SDNs) induce a long failure recovery process because of frequent flow operations between the SDN controller and switches. Accordingly, it is indispensable to design a new resilience approach to balance failure recovery time and forwarding rule occupation. To this end, we leverage flexible flow aggregation in fast reroute to solve this problem. In the proposed approach, each disrupted traffic flow is reassigned to a local reroute path for the purpose of congestion avoidance. Thus, all traffic flows assigned to the same local reroute path are aggregated into a new “big” flow, and the number of reconfigured forwarding rules in the restoration process is greatly reduced. We first formulate this problem as an integer linear programming model, then design an efficient heuristic named the “congestion-aware local fast reroute” (CALFR). Extensive emulation results show that CALFR enables fast recovery while avoiding link congestion in the post-recovery network.

An Entropy-Based DDoS Defense Mechanism in Software Defined Networks

The issue on defensing against Distributed Denial of Service (DDoS) attacks in Software Defined Networks (SDN) has been highly concerned by academe and industry. The existing studies cannot eliminate the false positives by using the simple classification algorithms. In this paper, we analyze the essential difference between DDoS attacks and flash crowds which causes some similar consequences to DDoS. Accordingly we design a novel effective Entropy-based DDoS Defense Mechanism (EDDM) running on the SDN controller, which including a two-stage DDoS detection method. Compared with the existing works, the EDDM avoids the dropping of legitimate packets and minimizes the losses of legitimate users. Simulations demonstrate that the EDDM can distinguish the DDoS attacks from flash crowds, find the locations of bots, and block attack packets at source effectively.