Showing papers on "Load balancing (computing) published in 2012"

Reduze 2 - Distributed Feynman Integral Reduction

Abstract: Reduze is a computer program for reducing Feynman integrals to master integrals employing a variant of Laporta's reduction algorithm. This article describes version 2 of the program. New features include the distributed reduction of single topologies on multiple processor cores. The parallel reduction of different topologies is supported via a modular, load balancing job system. Fast graph and matroid based algorithms allow for the identification of equivalent topologies and integrals.

Stochastic models of load balancing and scheduling in cloud computing clusters

Abstract: Cloud computing services are becoming ubiquitous, and are starting to serve as the primary source of computing power for both enterprises and personal computing applications. We consider a stochastic model of a cloud computing cluster, where jobs arrive according to a stochastic process and request virtual machines (VMs), which are specified in terms of resources such as CPU, memory and storage space. While there are many design issues associated with such systems, here we focus only on resource allocation problems, such as the design of algorithms for load balancing among servers, and algorithms for scheduling VM configurations. Given our model of a cloud, we first define its capacity, i.e., the maximum rates at which jobs can be processed in such a system. Then, we show that the widely-used Best-Fit scheduling algorithm is not throughput-optimal, and present alternatives which achieve any arbitrary fraction of the capacity region of the cloud. We then study the delay performance of these alternative algorithms through simulations.

Logically centralized?: state distribution trade-offs in software defined networks

Abstract: Software Defined Networks (SDN) give network designers freedom to refactor the network control plane. One core benefit of SDN is that it enables the network control logic to be designed and operated on a global network view, as though it were a centralized application, rather than a distributed system - logically centralized. Regardless of this abstraction, control plane state and logic must inevitably be physically distributed to achieve responsiveness, reliability, and scalability goals. Consequently, we ask: "How does distributed SDN state impact the performance of a logically centralized control application?"Motivated by this question, we characterize the state exchange points in a distributed SDN control plane and identify two key state distribution trade-offs. We simulate these exchange points in the context of an existing SDN load balancer application. We evaluate the impact of inconsistent global network view on load balancer performance and compare different state management approaches. Our results suggest that SDN control state inconsistency significantly degrades performance of logically centralized control applications agnostic to the underlying state distribution.

Online team formation in social networks

Abstract: We study the problem of online team formation. We consider a setting in which people possess different skills and compatibility among potential team members is modeled by a social network. A sequence of tasks arrives in an online fashion, and each task requires a specific set of skills. The goal is to form a new team upon arrival of each task, so that (i) each team possesses all skills required by the task, (ii) each team has small communication overhead, and (iii) the workload of performing the tasks is balanced among people in the fairest possible way.We propose efficient algorithms that address all these requirements: our algorithms form teams that always satisfy the required skills, provide approximation guarantees with respect to team communication overhead, and they are online-competitive with respect to load balancing. Experiments performed on collaboration networks among film actors and scientists, confirm that our algorithms are successful at balancing these conflicting requirements.This is the first paper that simultaneously addresses all these aspects. Previous work has either focused on minimizing coordination for a single task or balancing the workload neglecting coordination costs.

Online algorithms for geographical load balancing

Abstract: It has recently been proposed that Internet energy costs, both monetary and environmental, can be reduced by exploiting temporal variations and shifting processing to data centers located in regions where energy currently has low cost. Lightly loaded data centers can then turn off surplus servers. This paper studies online algorithms for determining the number of servers to leave on in each data center, and then uses these algorithms to study the environmental potential of geographical load balancing (GLB). A commonly suggested algorithm for this setting is “receding horizon control” (RHC), which computes the provisioning for the current time by optimizing over a window of predicted future loads. We show that RHC performs well in a homogeneous setting, in which all servers can serve all jobs equally well; however, we also prove that differences in propagation delays, servers, and electricity prices can cause RHC perform badly, So, we introduce variants of RHC that are guaranteed to perform as well in the face of such heterogeneity. These algorithms are then used to study the feasibility of powering a continent-wide set of data centers mostly by renewable sources, and to understand what portfolio of renewable energy is most effective.

Distributed $\alpha$ -Optimal User Association and Cell Load Balancing in Wireless Networks

Abstract: In this paper, we develop a framework for user association in infrastructure-based wireless networks, specifically focused on flow-level cell load balancing under spatially inhomogeneous traffic distributions. Our work encompasses several different user association policies: rate-optimal, throughput-optimal, delay-optimal, and load-equalizing, which we collectively denote α-optimal user association. We prove that the optimal load vector ρ* that minimizes a generalized system performance function is the fixed point of a certain mapping. Based on this mapping, we propose and analyze an iterative distributed user association policy that adapts to spatial traffic loads and converges to a globally optimal allocation. We then address admission control policies for the case where the system is overloaded. For an appropriate system-level cost function, the optimal admission control policy blocks all flows at cells edges. However, providing a minimum level of connectivity to all spatial locations might be desirable. To this end, a location-dependent random blocking and user association policy are proposed.

A Survey of Load Balancing in Cloud Computing: Challenges and Algorithms

Abstract: Load Balancing is essential for efficient operations indistributed environments. As Cloud Computing is growingrapidly and clients are demanding more services and betterresults, load balancing for the Cloud has become a veryinteresting and important research area. Many algorithms weresuggested to provide efficient mechanisms and algorithms forassigning the client's requests to available Cloud nodes. Theseapproaches aim to enhance the overall performance of the Cloudand provide the user more satisfying and efficient services. Inthis paper, we investigate the different algorithms proposed toresolve the issue of load balancing and task scheduling in CloudComputing. We discuss and compare these algorithms to providean overview of the latest approaches in the field.

Tarazu: optimizing MapReduce on heterogeneous clusters

Abstract: Data center-scale clusters are evolving towards heterogeneous hardware for power, cost, differentiated price-performance, and other reasons. MapReduce is a well-known programming model to process large amount of data on data center-scale clusters. Most MapReduce implementations have been designed and optimized for homogeneous clusters. Unfortunately, these implementations perform poorly on heterogeneous clusters (e.g., on a 90-node cluster that contains 10 Xeon-based servers and 80 Atom-based servers, Hadoop performs worse than on 10-node Xeon-only or 80-node Atom-only homogeneous sub-clusters for many of our benchmarks). This poor performance remains despite previously proposed optimizations related to management of straggler tasks. In this paper, we address MapReduce's poor performance on heterogeneous clusters. Our first contribution is that the poor performance is due to two key factors: (1) the non-intuitive effect that MapReduce's built-in load balancing results in excessive and bursty network communication during the Map phase, and (2) the intuitive effect that the heterogeneity amplifies load imbalance in the Reduce computation. Our second contribution is Tarazu, a suite of optimizations to improve MapReduce performance on heterogeneous clusters. Tarazu consists of (1) Communication-Aware Load Balancing of Map computation (CALB) across the nodes, (2) Communication-Aware Scheduling of Map computation (CAS) to avoid bursty network traffic and (3) Predictive Load Balancing of Reduce computation (PLB) across the nodes. Using the above 90-node cluster, we show that Tarazu significantly improves performance over a baseline of Hadoop with straightforward tuning for hardware heterogeneity.

Load Balancing of Nodes in Cloud Using Ant Colony Optimization

Abstract: In this paper, we proposed an algorithm for load distribution of workloads among nodes of a cloud by the use of Ant Colony Optimization (ACO). This is a modified approach of ant colony optimization that has been applied from the perspective of cloud or grid network systems with the main aim of load balancing of nodes. This modified algorithm has an edge over the original approach in which each ant build their own individual result set and it is later on built into a complete solution. However, in our approach the ants continuously update a single result set rather than updating their own result set. Further, as we know that a cloud is the collection of many nodes, which can support various types of application that is used by the clients on a basis of pay per use. Therefore, the system, which is incurring a cost for the user should function smoothly and should have algorithms that can continue the proper system functioning even at peak usage hours.

An energy-aware distributed clustering protocol in wireless sensor networks using fuzzy logic

Abstract: Clustering is an effective approach for organizing a network into a connected hierarchy, load balancing, and prolonging the network lifetime. On the other hand, fuzzy logic is capable of wisely blending different parameters. This paper proposes an energy-aware distributed dynamic clustering protocol (ECPF) which applies three techniques: (1) non-probabilistic cluster head (CH) elections, (2) fuzzy logic, and (3) on demand clustering. The remaining energy of the nodes is the primary parameter for electing tentative CHs via a non-probabilistic fashion. A non-probabilistic CH election is implemented by introducing a delay inversely proportional to the residual energy of each node. Therefore, tentative CHs are selected based on their remaining energy. In addition, fuzzy logic is employed to evaluate the fitness (cost) of a node in order to choose a final CH from the set of neighboring tentative CHs. On the other hand, every regular (non CH) node elects to connect to the CH with the least fuzzy cost in its neighborhood. Besides, in ECPF, CH elections are performed sporadically (in contrast to performing it every round). Simulation results demonstrate that our approach performs better than well known protocols (LEACH, HEED, and CHEF) in terms of extending network lifetime and saving energy.

Cloud Load Balancing Techniques : A Step Towards Green Computing

Abstract: Cloud computing is emerging as a new paradigm of large-scale distributed computing. It is a framework for enabling convenient, on-demand network access to a shared pool of computing resources. Load balancing is one of the main challenges in cloud computing which is required to distribute the dynamic workload across multiple nodes to ensure that no single node is overwhelmed. It helps in optimal utilization of resources and hence in enhancing the performance of the system. The goal of load balancing is to minimize the resource consumption which will further reduce energy consumption and carbon emission rate that is the dire need of cloud computing. This determines the need of new metrics, energy consumption and carbon emission for energy-efficient load balancing in cloud computing. This paper discusses the existing load balancing techniques in cloud computing and further compares them based on various parameters like performance, scalability, associated overhead etc. that are considered in different techniques. It further discusses these techniques from energy consumption and carbon emission perspective.

Energy-Aware Autonomic Resource Allocation in Multitier Virtualized Environments

Abstract: With the increase of energy consumption associated with IT infrastructures, energy management is becoming a priority in the design and operation of complex service-based systems. At the same time, service providers need to comply with Service Level Agreement (SLA) contracts which determine the revenues and penalties on the basis of the achieved performance level. This paper focuses on the resource allocation problem in multitier virtualized systems with the goal of maximizing the SLAs revenue while minimizing energy costs. The main novelty of our approach is to address-in a unifying framework-service centers resource management by exploiting as actuation mechanisms allocation of virtual machines (VMs) to servers, load balancing, capacity allocation, server power state tuning, and dynamic voltage/frequency scaling. Resource management is modeled as an NP-hard mixed integer nonlinear programming problem, and solved by a local search procedure. To validate its effectiveness, the proposed model is compared to top-performing state-of-the-art techniques. The evaluation is based on simulation and on real experiments performed in a prototype environment. Synthetic as well as realistic workloads and a number of different scenarios of interest are considered. Results show that we are able to yield significant revenue gains for the provider when compared to alternative methods (up to 45 percent). Moreover, solutions are robust to service time and workload variations.

Ant colony Optimization: A Solution of Load balancing in Cloud

Abstract: As the cloud computing is a new style of computing over internet. It has many advantages along with some crucial issues to be resolved in order to improve reliability of cloud environment. These issues are related with the load management, fault tolerance and different security issues in cloud environment. In this paper the main concern is load balancing in cloud computing. The load can be CPU load, memory capacity, delay or network load. Load balancing is the process of distributing the load among various nodes of a distributed system to improve both resource utilization and job response time while also avoiding a situation where some of the nodes are heavily loaded while other nodes are idle or doing very little work. Load balancing ensures that all the processor in the system or every node in the network does approximately the equal amount of work at any instant of time. Many methods to resolve this problem has been came into existence like Particle Swarm Optimization, hash method, genetic algorithms and several scheduling based algorithms are there. In this paper we are proposing a method based on Ant Colony optimization to resolve the problem of load balancing in cloud environment.

Asymptotic independence of queues under randomized load balancing

Abstract: Randomized load balancing greatly improves the sharing of resources while being simple to implement. In one such model, jobs arrive according to a rate-?N Poisson process, with ?<1, in a system of N rate-1 exponential server queues. In Vvedenskaya et al. (Probl. Inf. Transm. 32:15---29, 1996), it was shown that when each arriving job is assigned to the shortest of D, D?2, randomly chosen queues, the equilibrium queue sizes decay doubly exponentially in the limit as N??. This is a substantial improvement over the case D=1, where queue sizes decay exponentially. The reasoning in Vvedenskaya et al. (Probl. Inf. Transm. 32:15---29, 1996) does not easily generalize to jobs with nonexponential service time distributions. A modularized program for treating randomized load balancing problems with general service time distributions was introduced in Bramson et al. (Proc. ACM SIGMETRICS, pp. 275---286, 2010). The program relies on an ansatz that asserts that, for a randomized load balancing scheme in equilibrium, any fixed number of queues become independent of one another as N??. This allows computation of queue size distributions and other performance measures of interest. In this article, we demonstrate the ansatz in several settings. We consider the least loaded balancing problem, where an arriving job is assigned to the queue with the smallest workload. We also consider the more difficult problem, where an arriving job is assigned to the queue with the fewest jobs, and demonstrate the ansatz when the service discipline is FIFO and the service time distribution has a decreasing hazard rate. Last, we show the ansatz always holds for a sufficiently small arrival rate, as long as the service distribution has 2 moments.

Network storage systems having clustered raids for improved redundancy and load balancing

Abstract: A clustered network-based storage system includes a host server, multiple high availability system controller pairs, and multiple storage devices across multiple arrays. Two independent storage array subsystems each include a quorum drive copy and are each controlled by a HA pair, with remote volume mirroring links coupling the separate HA pairs. The host server includes a virtualization agent that identifies and prioritizes communication paths, and also determines capacity across all system nodes. A system storage management agent determines an overall storage profile across the system. The virtualization agent, storage management agent, quorum drive copies and remote volume mirroring link all operate to provide increased redundancy, load sharing, or both between the separate first and second arrays of storage devices.

Load Balancing for MapReduce-based Entity Resolution

Abstract: The effectiveness and scalability of MapReduce-based implementations of complex data-intensive tasks depend on an even redistribution of data between map and reduce tasks. In the presence of skewed data, sophisticated redistribution approaches thus become necessary to achieve load balancing among all reduce tasks to be executed in parallel. For the complex problem of entity resolution, we propose and evaluate two approaches for such skew handling and load balancing. The approaches support blocking techniques to reduce the search space of entity resolution, utilize a preprocessing MapReduce job to analyze the data distribution, and distribute the entities of large blocks among multiple reduce tasks. The evaluation on a real cloud infrastructure shows the value and effectiveness of the proposed load balancing approaches.

Improved Max-Min Algorithm in Cloud Computing

Abstract: paper, a unique modification of Max-min algorithm is proposed. The algorithm is built based on comprehensive study of the impact of RASA algorithm in scheduling tasks and the atom concept of Max-min strategy. An Improved version of Max-min algorithm is proposed to outperform scheduling map at least similar to RASA map in total complete time for submitted jobs. Improved Max-min is based on the expected execution time instead of complete time as a selection basis. Experimental results show availability of load balance in small cloud computing environment and total small makespan in large-scale distributed system; cloud computing. In turn scheduling tasks within cloud computing using Improved Max-min demonstrates achieving schedules with comparable lower makespan rather than RASA and original Max-min.

A New Resource Scheduling Strategy Based on Genetic Algorithm in Cloud Computing Environment

Abstract: In view of the load balancing problem in VM resources scheduling, this paper presents a scheduling strategy on load balancing of VM resources based on genetic algorithm. According to historical data and current state of the system and through genetic algorithm, this strategy computes ahead the influence it will have on the system after the deployment of the needed VM resources and then chooses the least-affective solution, through which it achieves the best load balancing and reduces or avoids dynamic migration. At the same time, this paper brings in variation rate to describe the load variation of system virtual machines, and it also introduces average load distance to measure the overall load balancing effect of the algorithm. The experiment shows that this strategy has fairly good global astringency and efficiency, and the algorithm of this paper is, to a great extent, able to solve the problems of load imbalance and high migration cost after system VM being scheduled. What is more, the average load distance does not grow with the increase of VM load variation rate, and the system scheduling algorithm has quite good resource utility.

Dynamic cell association for downlink sum rate maximization in multi-cell heterogeneous networks

Abstract: In this work, we consider a heterogeneous network consisting in several macro nodes and pico nodes. Our goal is to associate users, belonging to this network, to one of the nodes, while maximizing the sum rate of all users. We also want to analyze the load balancing achieved by this association. Therefore, we develop a new theoretical framework to study cell association for the downlink of multi-cell networks and derive an upper bound on the achievable sum rate. We propose a dynamic cell association heuristic, which achieves performance close to optimal. Finally, we verify our results through numerical evaluations and implement the proposed heuristic in an LTE simulator to demonstrate its viability.

Energy-aware load balancing in content delivery networks

Abstract: Internet-scale distributed systems such as content delivery networks (CDNs) operate hundreds of thousands of servers deployed in thousands of data center locations around the globe. Since the energy costs of operating such a large IT infrastructure are a significant fraction of the total operating costs, we argue for redesigning CDNs to incorporate energy optimizations as a first-order principle. We propose techniques to turn off CDN servers during periods of low load while seeking to balance three key design goals: maximize energy reduction, minimize the impact on client-perceived service availability (SLAs), and limit the frequency of on-off server transitions to reduce wear-and-tear and its impact on hardware reliability. We propose an optimal offline algorithm and an online algorithm to extract energy savings both at the level of local load balancing within a data center and global load balancing across data centers. We evaluate our algorithms using real production workload traces from a large commercial CDN. Our results show that it is possible to reduce the energy consumption of a CDN by 51% while ensuring a high level of availability that meets customer SLA requirements and incurring an average of one on-off transition per server per day. Further, we show that keeping even 10% of the servers as hot spares helps absorb load spikes due to global flash crowds and minimize any impact on availability SLAs. Finally, we show that redistributing load across highly proximal data centers can enhance service availability significantly, but has only a modest impact on energy savings.

Load balancing cloud computing: State of art

Abstract: The rapid development of Internet has given birth to a new business model: Cloud Computing. This new paradigm has experienced a fantastic rise in recent years. Because of its infancy, it remains a model to be developed. In particular, it must offer the same features of services than traditional systems. The cloud computing is large distributed systems that employ distributed resources to deliver a service to end users by implementing several technologies. Hence providing acceptable response time for end users, presents a major challenge for cloud computing. All components must cooperate to meet this challenge, in particular through load balancing algorithms. This will enhance the availability and will gain the end user confidence. In this paper we try to give an overview of load balancing in the cloud computing by exposing the most important research challenges.

BalanceFlow: Controller load balancing for OpenFlow networks

Abstract: In the discussion about Future Internet, Software-Defined Networking (SDN), enabled by OpenFlow, is currently seen as one of the most promising paradigm While the availability and scalability concerns rises as a single controller could be alleviated by using replicate or distributed controllers, there lacks a flexible mechanism to allow controller load balancing This paper proposes BalanceFlow, a controller load balancing architecture for OpenFlow networks By utilizing CONTROLLER X action extension for OpenFlow switches and cross-controller communication, one of the controllers, called “super controller”, can flexibly tune the flow-requests handled by each controller, without introducing unacceptable propagation latencies Experiments based on real topology show that BalanceFlow can adjust the load of each controller dynamically

Load Balancing in MapReduce Based on Scalable Cardinality Estimates

Abstract: MapReduce has emerged as a popular tool for distributed and scalable processing of massive data sets and is being used increasingly in e-science applications. Unfortunately, the performance of MapReduce systems strongly depends on an even data distribution while scientific data sets are often highly skewed. The resulting load imbalance, which raises the processing time, is even amplified by high runtime complexity of the reducer tasks. An adaptive load balancing strategy is required for appropriate skew handling. In this paper, we address the problem of estimating the cost of the tasks that are distributed to the reducers based on a given cost model. An accurate cost estimation is the basis for adaptive load balancing algorithms and requires to gather statistics from the mappers. This is challenging: (a) Since the statistics from all mappers must be integrated, the mapper statistics must be small. (b) Although each mapper sees only a small fraction of the data, the integrated statistics must capture the global data distribution. (c) The mappers terminate after sending the statistics to the controller, and no second round is possible. Our solution to these challenges consists of two components. First, a monitoring component executed on every mapper captures the local data distribution and identifies its most relevant subset for cost estimation. Second, an integration component aggregates these subsets approximating the global data distribution.

Delay-based congestion control for multipath TCP

Abstract: With the aid of multipath transport protocols, a multihomed host can shift some of its traffic from more congested paths to less congested ones, thus compensating for lost bandwidth on some paths by moderately increasing transmission rates on other ones. However, existing multipath proposals achieve only coarse-grained load balancing due to a rough estimate of network congestion using packet losses. This paper formulates the problem of multipath congestion control and proposes an approximate iterative algorithm to solve it. We prove that a fair and efficient traffic shifting implies that every flow strives to equalize the extent of congestion that it perceives on all its available paths.We call this result “Congestion Equality Principle”. By instantiating the approximate iterative algorithm, we develop weighted Vegas (wVegas), a delay-based algorithm for multipath congestion control, which uses packet queuing delay as congestion signals, thus achieving fine-grained load balancing. Our simulations show that, compared with loss-based algorithms, wVegas is more sensitive to changes of network congestion and thus achieves more timely traffic shifting and quicker convergence. Additionally, as it occupies fewer link buffers, wVegas rarely causes packet losses and shows better intra-protocol fairness.

Execution analysis of load balancing algorithms in cloud c omputing environment

Abstract: The concept oft Cloud computing has significantly changed the field of parallel and distributed computing systems today. Cloud computing enables a wide range of users to access distributed, scalable, virtualized hardware and/or software infrastructure over the Internet. Load balancing is a methodology to distribute workload across multiple computers, or other resources over the network links to achieve optimal resource utilization, maximize throughput, minimum response time, and avoid overload. With recent advent of technology, resource control or load balancing in cloud computing is main challenging issue. A few existing scheduling algorithms can maintain load balancing and provide better strategies through efficient job scheduling and resource allocation techniques as well. In order to gain maximum profits with optimized load balancing algorithms, it is necessary to utilize resources efficiently. This paper presents a review of a few load balancing algorithms or technique in cloud computing. The objective of this paper is to identify qualitative components for simulation in cloud environment and then based on these components, execution analysis of load balancing algorithms are also presented.

xOMB: extensible open middleboxes with commodity servers

Abstract: This paper presents the design and implementation of an incrementally scalable architecture for middleboxes based on commodity servers and operating systems. xOMB, the eXtensible Open MiddleBox, employs general programmable network processing pipelines, with user-defined C++ modules responsible for parsing, transforming, and forwarding network flows. We implement three processing pipelines in xOMB, demonstrating good performance for load balancing, protocol acceleration, and application integration. In particular, our xOMB load balancing switch is able to match or outperform a commercial programmable switch and popular open-source reverse proxy while still providing a more flexible programming model.

Load-balancing via modulus distribution and tcp flow redirection due to server overload

Abstract: A method, switch, and/or computer program product routes IP packet flows. An Ethernet switch receives an IP packet flow. Each of the packets in the IP packet flow has a header that contains a same 5-tuple. A load balancing control engine determines whether servers in a group of servers are balanced in their utilization according to 5-tuple redirection rules contained in the load balancing control engine. In response to the load balancing control engine determining, according to the 5-tuple redirection rules, that the servers are balanced, the Ethernet switch routes the IP packet flow to the servers. In response to the load balancing control engine determining that the servers are unbalanced, the load balancing control engine instructs the Ethernet switch to redirect the IP packet flow to a server that is relatively less busy than other servers.

Method and apparatus for load balancing

Abstract: A method and apparatus is provided in which a multi-site load balancing system directs a client to a site best able to respond to the client's request based on a combination of balancing methods. Performance metric balancing is employed to select sites having the best performance metrics to participate in network latency balancing to determine the site best able to respond the request. The sites participating in the network latency balancing are selected based on having performance metrics within an allowable deviation of the best performance metric. Alternatively, network latency balancing is employed to select sites having the least network latency to participate in performance metric balancing to determine the site best able to respond to the request.

Walk the line: consistent network updates with bandwidth guarantees

Abstract: New advances in technologies for high-speed and seamless migration of VMs turns VM migration into a promising and efficient means for load balancing, configuration, power saving, attaining a better resource utilization by reallocating VMs, cost management, etc. in data centers. Despite these numerous benefits, VM migration is still a challenging task for providers, since moving VMs requires update of network state, which consequently could lead to inconsistencies, outages, creation of loops and violations of service level (SLA) agreement requirements. Many applications today like financial services, social networking, recommendation systems, and web search cannot tolerate such problems or degradation of service [5, 12]. On the positive side, the emerging trend of Software Defined Networking (SDN) provides a powerful tool for tackling these challenging problems. In SDN, management applications are run by a logically-centralized controller that directly controls the packet handling functionality of the underlying switches. For example, OpenFlow, a recently proposed mechanism for SDN, provides an API that allows the controller to install rules in switches, process data packets, learn the topology changes, and query traffic counters [13]. The ability to run algorithms in a logically centralized location, and precisely manipulate the forwarding layer of switches creates a new opportunity for transitioning the network between two states. In particular this paper studies the question: given a start-