Task Scheduling and Lightpath Establishment in Optical Grids

Data-intensive Grid applications require huge data transferring between multiple geographically separated computing nodes where computing tasks are executed. For a future WDM network to efficiently support this type of emerging applications, neither the traditional approaches to establishing lightpaths between given source destination pairs are sufficient, nor are those existing application level approaches that consider computing resources but ignore the optical layer connectivity. Instead, lightpath establishment has to be considered jointly with task scheduling to achieve best performance. In this paper, we study the optimization problems of jointly scheduling both computing resources and network resources. We first present the formulation of two optimization problems with the objectives being the minimization of the completion time of a job and minimization of the resource usage/cost to satisfy a job with a deadline. When the objective is to minimize the completion time, we devise an optimal algorithm for a special type of applications. Furthermore, we propose efficient heuristics to deal with general applications with either optimization objective and demonstrate their good performances in simulation.     

Edge-Reconfigurable Optical Networks (ERONs): Rationale, Network Design, and Evaluation

To bridge the gap between the current practice of setting up expensive, dedicated, lightpath connections (i.e., static topologies), and the distant future vision of inexpensive access to dynamically switched end-to-end lightpaths, we propose a medium term solution in the form of edge-reconfigurable optical networks (ERONs). An ERON is an overlay-control network created by installing readily available MEMS optical switches, and implementing a GMPLS control plane at sites interconnected by static lightpaths. The switches and control software are deployed at the edge of the network and operated by the organization-user (i.e., outside the network provider's control), hence the term ldquoedge-reconfigurablerdquo. By providing dynamic, automated control of end-to-end lightpaths, ERONs enable the sharing of expensive network resources among multiple users and applications that require sporadic access to these resources. We develop an algorithm for creating an ERON from an existing topology of static lightpaths. We also present simulation results that quantify the benefits of ERONs, in terms of the number of lightpaths that are needed when compared to a static configuration of independent and dedicated circuits.     

Reconfiguration of Logical Topologies with Minimum Traffic Disruptions in Reliable WDM-Based Mesh Networks

A wavelength division multiplexing (WDM) network offers a flexible networking infrastructure by assigning the route and wavelength of lightpaths. We can construct an optimal logical topology, by properly setting up the lightpaths. Furthermore, setting up a backup lightpath for each lightpath improves network reliability. When traffic demand changes, a new optimal (or sub-optimal) topology should be obtained by again applying the formulation. Then, we can reconfigure the running topology to the logical topology obtained. However, during this reconfiguration, traffic loss may occur due to the deletion of older lightpaths. In this paper, we consider reconfiguring the logical topology in reliable WDM-based mesh networks, and we propose five procedures that can be used to reconfigure a running lightpath to a new one. Applying the procedures one by one produces a new logical topology. The procedures mainly focus on utilizing free wavelength resources and the resources of backup lightpaths, which are not used usually for transporting traffic. The results of computer simulations indicate that the traffic loss is remarkably reduced in the 14-node network we used as an example.     

Dynamic circuit provisioning in all-optical WDM networks using lightpath switching

In this paper we investigate the problem of provisioning holding-time-aware (HTA) dynamic circuits in all-optical wavelength division multiplexed (WDM) networks. We employ a technique called lightpath switching (LPS) wherein the data transmission may begin on one lightpath and switch to a different lightpath at a later time. Lightpath switches are transparent to the user and are managed by the network. Allowing LPS creates a number of segments that can use independent lightpaths. We first compare the performance of traditional routing and wavelength (RWA) assignment to routing and wavelength assignment with LPS. We show that LPS can significantly reduce blocking compared to traditional RWA. We then address the problem of routing dynamic anycast HTA dynamic circuits. We propose two heuristics to solve the anycast RWA problem: anycast with continuous segment (ACS) and anycast with lightpath switching (ALPS). In ALPS we exercise LPS, and provision a connection request by searching for the best candidate destination node is such a way that the network resources are utilized efficiently. In ACS we do not allow a connection request to switch lightpaths. The lightpaths to each candidate destination node of a request are computed using traditional RWA algorithms. We first compare the performance of ACS to ALPS and observe that ALPS achieves better blocking than ACS. Furthermore, we also compare the performance of these two anycast RWA algorithms to the traditional unicast RWA algorithm. We show that the anycast RWA algorithms presented here significantly outperform the traditional unicast RWA algorithms.     

Idle Protection Capacity Reuse in Multiclass Optical Networks

Optical networks carrying traffic belonging to different survivability classes must ensure not only the proper survivability differentiation but also the efficient network resource utilization. Current approaches improve network resource utilization by carrying low-class preemptable traffic along high-class spare resources, which are idle (i.e., unutilized) during failure-free conditions. In this paper, the proposed idle protection capacity reuse (IR) framework generalizes the practice of reutilizing idle spare resources to both failure-free and failure conditions. The IR framework is based on the idea of exploiting idle high-class lightpath protection resources not only for routing low-class lightpaths during the provisioning phase (i.e., provisioning-phase IR, P-IR) but also for dynamically restoring low-class lightpaths during the recovery phase (i.e., recovery-phase IR, R-IR). Both P-IR and R-IR have the potential to improve network utilization while providing multiclass lightpaths with the required survivability differentiation. The numerical evaluation shows that the utilization of R-IR in dynamic restoration results in an improved low-class lightpath survivability with respect to the utilization of stub release. Moreover, if P-IR and R-IR are alternatively exploited, then the former improves the provisioning performance, while the latter guarantees a higher survivability. In the end, if P-IR and R-IR are concurrently employed in the provisioning and in the dynamic restoration of two different low classes of lightpaths, respectively, an inherent survivability differentiation is achieved     

Efficient Protection in MPλS Networks Using Backup Trees: Part One—Concepts and Heuristics

Multi-protocol lambda switching (MPS) has recently been applied in the optical network control plane to provide fast lightpath provisioning. As an increasing amount of traffic is carried in optical transport networks (OTNs), single network failures can affect a vast amount of traffic, making lightpath protection crucial. Therefore, shared backup tree (BT) lightpath protection is a promising paradigm in MPS networks due to its ability of fast recovery and its efficiency in consumed resources. A shared BT is used to protect a group of working lightpaths towards the same destination. From the working lightpaths in such a group, only one affected lightpath at a time can be recovered using the BT. The main problem is how to group and route the working paths (WPs) and how to route the BTs, in such a way that the capacity resources used by the WPs and the BTs are minimized. In Part One of this study (presented in this paper), we propose three approaches to cope with this problem. The first approach is a purely integer linear programming (ILP) based method. The second one is a combination of ILP and a heuristic technique. The last one is a purely heuristic approach. In this paper, these approaches are theoretically compared. In Part Two [1] of this study, several simulations are carried out in order to compare these approaches in terms of performance and computing effort. The experimental results are in line with the theoretical expectations.     

Post‐Disaster least loaded lightpath routing in elastic optical networks

Disaster events directly affect the physical topology of core networks and may lead to simultaneous failure of multiple lightpaths leading to massive service outages for network operators. To recover from such a failure scenario, the existing routing algorithms running on network nodes (routers or switches) typically attempt to reestablish the connections over new routes with shortest distances and hop count approach. However, this approach may result in congestion on some links, while other links may have the unutilized capacity. Hence, intelligent lightpath computing techniques are required to efficiently route network traffic over the new routes by considering traffic load of each link in addition to distance and hop count to minimize network congestion. In this paper, we have proposed a capacity‐constrained maximally spatial disjoint lightpath algorithm to tackle the provisioning and restoration of disrupted lightpaths in a postdisaster scenario in the context of elastic optical networking. This algorithm computes an alternate least loaded lightpath for disrupted primary lightpath using capacity‐constrained shortest lightpath. Alternate lightpath selection is based on a criteria parameter for a lightpath to be least loaded and constrained by either the length or the spatial distance between primary and alternate lightpaths. The spatial distance between lightpaths enables to reestablish the disrupted connection request away from disaster proximity. The performance of the proposed algorithm is evaluated through simulation for several parameters like blocking probability, network utilization, connection success rates, and minimum spatial distance.     

ASON光网格中基于GridFTP的资源并行调度机制

为解决面向应用驱动的网格资源动态分配,提出基于ASON的光网格模型,在此基础上提出面向GridFTP服务的并行网络资源调度机制.并行资源调度借助于光资源控制系统提供网格应用中并行通道数据传输,在网格应用驱动下实现网络资源的动态按需分配,提高了光资源利用效率及海量数据传输能力.通过构建基于ASON的光网格平台体系,执行DridFTP文件传输应用服务以验证所提模型和资源调度机制.     

Sliding scheduled lightpath provisioning by mixed partition coloring in WDM optical networks

In WDM optical networks, lightpath provisioning for static, incremental and dynamic traffic model has been widely investigated. However, Internet connectivity services are increasingly showing a new kind of traffic type in the context of optical networks, i.e., sliding scheduled traffic, which does not have a rigid deadline and allows flexible sliding within a large time window. This new traffic type offers opportunity of more efficiently utilizing network resources to accommodate more traffic, and poses new challenges of exploiting the flexibility of scheduling time. In this paper, we formulate the static sliding scheduled lightpath demand (SSLD) provisioning problem as a mixed partition coloring model in which routing and wavelength assignment are conducted simultaneously in compliance with the allowed time window of each request. Then, we propose a novel one-step heuristic algorithm named as maximum conflict degree first conflict reducing (MCDF-CR) to solve the SSLD provisioning problem based on mixed partition coloring model. Simulation results show that our approach can improve wavelengths utilization compared to previous heuristics.     

Logical topology design for dynamic traffic grooming in WDM optical networks

Traffic grooming in optical networks refers to consolidation of subwavelength client connections onto lightpaths. Depending on whether client connections are given in advance or randomly arrive/depart, traffic grooming is classified as static and dynamic. Dynamic traffic grooming has been traditionally performed through establishing/releasing lightpaths online. In this paper, the authors propose an alternate approach to design a static logical topology a priori and then route randomly arriving client connections on it to avoid frequent lightpath setup/teardown. Two problems are considered: 1) minimize resource usage constrained by traffic blocking requirements and 2) maximize performance constrained by given resources. These are formulated as integer linear-programming (ILP) problems. The numerical results show that the resource usage dramatically decreases when the blocking requirement is relaxed, and the grooming performance slowly increases when given more resources. In addition, the number of ports at client nodes has more profound impact on traffic grooming than the number of wavelengths.     

Comprehensive design methodology for control and data planes in wavelength-routed optical networks

We propose a comprehensive design methodology for control and data planes of wavelength-routed optical networks (WRONs) employing mixed-line-rate (MLR) transmission for cost-effective resource provisioning. The proposed approach attempts to minimize the maximum lightpath capacity demand in Gbps (representing the measure of lightpath congestion) in network for a given traffic matrix by using a mix of a heuristic scheme and linear programming (LP). In the first step of the proposed three-step design, some lightpaths are set up on a set of judiciously selected fiber links (with point-to-point lightpaths between neighboring nodes), on a specific wavelength throughout the network, and an appropriate fraction of the same set of lightpaths is utilized for carrying control information, forming therefore the control plane (CP) of the WRON. The remaining bandwidth of these lightpaths is utilized to carry the data traffic along with all other designed lightpaths of the WRON using appropriate algorithm, forming the overall data plane (DP) of the WRON. In the second step, traffic routing is carried out through LP to minimize lightpath congestion in the network. In the third step, we utilize the results of LP to assign rates to lightpaths, such that the cost (considering only the transceiver cost) of the network is minimized. This design leads to congestion-aware MLR network with due consideration to cost-effectiveness without compromising the network restoration response against link failures. We carry out simulation studies employing possible CPs using both symmetric (CP topology being same as the physical topology) as well as asymmetric (using fewer fiber links than the symmetric case) topology. The results of our simulations indicate that the proposed design of CP with symmetric/asymmetric topology and in-band transmission with sub-lightpath capacity can bring down network congestion and cost with respect to symmetric out-of-band transmission (using fully reserved lightpaths for CP), without any perceptible sacrifice in respect of the network restoration time. Failure can occur either in CP or DP, or in both the planes. We investigate the effect of design of CP with symmetric/asymmetric topology on network restoration time for single- and double-link failures. We further present DP design methodology with hybrid restoration scheme, i.e., combination of dedicated (1:1) path protection and path restoration. We analyze the effect of symmetric CP topology and degree of protection on the congestion of the network. Some lightpaths, that support more traffic, are protected against failures, while the others are left for path restoration in the event of failures. As more lightpaths are protected, the congestion and power consumption of network increase. We provide an analysis of the factors that come into play while altering the degree of protection and observe how the choice for the degree of protection in DP can be arrived at using an appropriate design methodology.     

Wavelength routing with spare reconfiguration for all-optical WDM networks

This study presents a wavelength-routing scheme with spare reconfiguration (SR) to construct dependable all-optical wavelength-division-multiplexing (WDM) networks. Path protection using shared spare lightpaths is a general wavelength-routing method for reducing blocking probability while minimizing demand for spare resources. However, in a dynamic traffic environment, this method may still yield a poor performance because a wavelength on a link is very likely to be continuously held by a spare lightpath and to be unable to be assigned to the working lightpath of a new connection. This study develops a spare reconfiguration mechanism with wavelength reassignment (SR/spl I.bar/WR) and path reassignment (SR/spl I.bar/PR) to make the spare dynamic and thus further reduce the blocking probability. The proposed wavelength routing with SR proceeds in three stages and has polynomial time complexity. Extensive simulation experiments were conducted on the NSFNET and the K5 fully connected network to investigate the performance of the proposed wavelength routing with SR. Results of this study show that the proposed wavelength routing with SR can reduce the blocking probability compared with the general wavelength routing with just shared spare lightpaths by choosing a positive tuning cost. In addition, the improvement of the blocking performance is maximized when using SR/spl I.bar/WR followed by SR/spl I.bar/PR.     

Label Preference Schemes for Lightpath Provisioning and Restoration in Distributed GMPLS Networks

In wavelength-routed networks based on a GMPLS control plane, the resource reservation protocol with traffic engineering extensions (RSVP-TE) allows to establish end-to-end lightpaths. The resource reservation can be blocked due to lack of available resources (forward blocking) or due to resource contentions (backward blocking). In wavelength-routed networks, the backward blocking is the predominant blocking contribution, when traffic load is low or highly-dynamic and when lightpath restoration takes place. To reduce the backward blocking, the paper proposes two label preference (LP) schemes compliant with RSVP-TE message exchanges. LP schemes provide the destination node with a label identifying the preferred wavelength to reserve. The preferred label is computed in a distributed way during the forward signaling phase, with the objective of assigning disjoint wavelengths to reservation attempts that may contend the resources. Simulation results demonstrate that, compared to other schemes, LP schemes are effective in reducing the backward blocking during both lightpath provisioning and restoration, without negatively impacting the forward blocking.     

Integrated Scheduling of Grid Applications in WDM Optical Light-Trail Networks

A novel task graph model, flexible task model (FTM), is proposed for modeling the grid computing tasks and the relationships among the tasks. In this model, a task may generate output before the task completes whereas previous work assumes that no output is available until the task is completed. In addition, a task in FTM can start to execute when it has collected a minimum amount of required input from its predecessors. FTM is more general and flexible than the conventional task graph model considered in previous work. Based on FTM, we investigate the problem of scheduling grid applications that integrates the resource allocation for task execution and service provisioning for subwavelength data communication between the tasks. Data communication between grid tasks under the FTM model is better supported using light-trails in wavelength division multiplexing (WDM) networks, than lightpaths. The objective is to minimize the total amount of time for task completion or makespan. Simulation results show that our proposed scheduling algorithm under FTM significantly reduces the total task completion time compared with that under the conventional task graph model. Moreover, the communication service provisioning using light-trails is very resource efficient.      

Rise time of silicon p-n-p photodiodes

Exact analytical expressions are derived for the short circuit photodiode currents excited by light pulses, under the assumption that the drift carrier velocity linearly depends on the electric field in the depletion layer. Reflection from the back surface of the photodiode is taken into account. Using the obtained expressions it is possible to establish a connection between the rise time t_rise and the product W_eff of the absorption coefficient (λ) and effective depletion layer width W_eff(W) at various ratios of the diode thickness and the effective depletion layer width. The influence of the RC-constant (where C is the photodiode effective capacity and R is the sum of the diode series and loading resistances) on the rise time is also analyzed.

One of the most important conclusions is that generally the rise time is larger for p-n-n⁺ photodiode configurations than for n-p-p⁺ configurations at the same substrate resistivity.     

Routing and dimensioning in optical networks under traffic growth models: an asymptotic approach

We consider the problem of routing and dimensioning in a large optical network where traffic is growing over time. A model of traffic in optical network lightpaths is presented. Lightpaths arrive randomly according to a time-varying Poisson process and hold for a random time with a general distribution. We propose a wavelength division multiplexing network that requires no capacity upgrading in a given time period T while allowing the operator to accommodate all the lightpath requests. We obtain an exact solution of the routing and dimensioning problem under an asymptotic regime where both the capacities and the arrival rates are large. For moderate link capacities, we propose a method to dimension the links so that the first lightpath request rejection occurs, with high probability, after the specified time T. This involves the computation of capacity-exhaustion probability $the probability that at least one lightpath request is rejected in the time period (0, T) due to lack of bandwidth/capacity on some link. Computation of the exact capacity-exhaustion probability is possible for a few specific holding time distributions (e.g., exponential). Since this requires large computing resources, it is feasible only for small networks. We propose a method to estimate the capacity-exhaustion probabilities for a large optical network with general holding time distribution based on the results of an asymptotic analysis. We show that this method has a low computational complexity and is quite accurate in the desired range of low capacity-exhaustion probabilities.     

Virtual-topology adaptation for WDM mesh networks under dynamic traffic

We present a new approach to the virtual-topology reconfiguration problem for a wavelength-division-multiplexing- based optical wide-area mesh network under dynamic traffic demand. By utilizing the measured Internet backbone traffic characteristics, we propose an adaptation mechanism to follow the changes in traffic without a priori knowledge of the future traffic pattern. Our work differs from most previous studies on this subject which redesign the virtual topology according to an expected (or known) traffic pattern, and then modify the connectivity to reach the target topology. The key idea of our approach is to adapt the underlying optical connectivity by measuring the actual traffic load on lightpaths continuously (periodically based on a measurement period) and reacting promptly to the load imbalances caused by fluctuations on the traffic, by either adding or deleting one or more lightpath at a time. When a load imbalance is encountered, it is corrected either by tearing down a lightpath that is lightly loaded or by setting up a new lightpath when congestion occurs. We introduce high and low watermark parameters on lightpath loads to detect any over- or underutilized lightpath, and to trigger an adaptation step. We formulate an optimization problem which determines whether or not to add or delete lightpaths at the end of a measurement period, one lightpath at a time, as well as which lightpath to add or delete. This optimization problem turns out to be a mixed-integer linear program. Simulation experiments employing the adaptation algorithm on realistic network scenarios reveal interesting effects of the various system parameters (high and low watermarks, length of the measurement period, etc.). Specifically, we find that this method adapts very well to the changes in the offered traffic.     

边缘算力网络中智能算力感知路由分配策略研究

增强现实、自动驾驶、智慧城市、工业互联网等新型业务应用对网络算力的需求逐渐增强,然而,边缘算力网络系统面临着网络共存的问题——负载不均衡,导致一部分边缘服务器无法满足业务应用的处理需求,另一部分边缘服务器的算力资源处于空闲状态.为了高效协同地感知利用泛在、异构的算力资源,提升6G通信网络的内生感知和算力自适应能力,急需...     

Switch failure in a two duplex unit standby system

A two duplex unit standby system with an imperfect switch and subject to preventive maintenance is discussed. It is assumed that switch is available at the time of need with probability p(= 1−q). The failure time of a duplex unit is taken to be negative exponential whereas the repair time distributions of duplex unit and switch and the time taken for preventive maintenance action, respectively, are assumed to be arbitrarily distributed. The analysis is carried out by using the regenerative point technique, and some particular cases are discussed.     

虚拟计算环境下面向应用的基于信任的资源匹配模型

为了解决虚拟计算环境中的资源合理调度问题,提出了一种基于信任的资源匹配模型--“资源滑动窗口”模型。首先对资源的静态属性进行分类,然后依据基于时间窗的贝叶斯信任模型对资源节点进行评价,同时考虑资源的负载,动态划分其实时性能。最后综合评估静态和动态属性,确定调度资源分配。该模型为不同任务和属性的资源调度策略奠定了基础,实现了“合适的资源服务于合适的任务”的目的。仿真实验表明所提的模型相对传统的调度算法,具有更高的系统任务执行成功率和资源利用率。