Evaluation of a Load Profiling Approach to Routing Guaranteed Bandwidth Flows

To support the diverse Quality of Service (QoS) requirements of real-time (e.g. audio/video) applications in integrated services networks, several routing algorithms that allow for the reservation of the needed bandwidth over a Virtual Circuit (VC) established on one of several candidate routes have been proposed. Traditionally, such routing is done using the least-loaded concept, and thus results in balancing the load across the set of candidate routes. In a recent study, we have established the inadequacy of this load balancing practice and proposed the use of load profiling as an alternative. Load profiling techniques allow the distribution of "available" bandwidth across a set of candidate routes to match the characteristics of incoming VC QoS requests. In this paper we thoroughly characterize the performance of VC routing using load profiling and contrast it to routing using load balancing and load packing. We do so both analytically and via extensive simulations of multi-class traffic routing in Virtual Path (VP) based networks. Our findings confirm that for routing guaranteed bandwidth flows in VP networks, load balancing is not desirable as it results in VP bandwidth fragmentation, which adversely affects the likelihood of accepting new VC requests. This fragmentation is more pronounced when the granularity of VC requests is large. Typically, this occurs when a common VC is established to carry the aggregate traffic flow of many high-bandwidth real-time sources. For VP-based networks, our simulation results show that our load-profiling VC routing scheme performs better or as well as the traditional load-balancing VC routing in terms of revenue under both skewed and uniform workloads. Furthermore, load-profiling routing improves routing fairness by proactively increasing the chances of admitting high-bandwidth connections.

Mesh partitioning: a multilevel balancing and refinement algorithm

Multilevel algorithms are a successful class of optimization techniques which addresses the mesh partitioning problem. They usually combine a graph contraction algorithm together with a local optimization method which refines the partition at each graph level. In this paper we present an enhancement of the technique which uses imbalance to achieve higher quality partitions. We also present a formulation of the Kernighan-Lin partition optimization algorithm which incorporates load-balancing. The resulting algorithm is tested against a different but related state-of-the-art partitioner and shown to provide improved results.

Load balance algorithms for anycast

Increasingly, replicated anycast servers are being used to deliver network applications and service ever increasing user requests. Therefore, the strategies used to guarantee network bandwidth prerequisites and perform load balancing across the nodes of an anycast group are critical to the performance of online applications. In this paper, we model user requests, network congestion and latency, and server load using a combination of hydro-dynamics and queuing theory to develop an efficient job distribution strategy. Current, anycast research does not explicitly consider the system load of nodes within an anycast groups when distributing requests. Therefore, the performance of a heavily loaded anycast system can quickly become congested and uneven as jobs are routed to closely linked nodes which are already saturated with requests. In comparison, the nodes of further away systems remain relatively unused because of other issues such as network bandwidth and latency during these times. Our system redirects requests from busy systems to the idle, remotely linked nodes, to process requests faster in spite of slower network access. Using an empirical study, we show this technique can improve request performance, and throughput with minimal network probing overhead.

Mesh partitioning: A multilevel balancing and refinement algorithm

Multilevel algorithms are a successful class of optimisation techniques which address the mesh partitioning problem. They usually combine a graph contraction algorithm together with a local optimisation method which refines the partition at each graph level. In this paper we present an enhancement of the technique which uses imbalance to achieve higher quality partitions. We also present a formulation of the Kernighan-Lin partition optimisation algorithm which incorporates load-balancing. The resulting algorithm is tested against a different but related state-of the-art partitioner and shown to provide improved results.

Quality balancing for parallel adaptive FEM

We present a dynamic distributed load balancing algorithm for parallel, adaptive finite element simulations using preconditioned conjugate gradient solvers based on domain-decomposition. The load balancer is designed to maintain good partition aspect ratios. It can calculate a balancing flow using different versions of diffusion and a variant of breadth first search. Elements to be migrated are chosen according to a cost function aiming at the optimization of subdomain shapes. We show how to use information from the second step to guide the first. Experimental results using Bramble's preconditioner and comparisons to existing state-ot-the-art load balancers show the benefits of the construction.

Modeling, stability analysis and control of microgrid

With the increase in the level of global warming, renewable energy based distributed generators (DGs) will increasingly play a dominant role in electricity production. Distributed generation based on solar energy (photovoltaic and solar thermal), wind, biomass, mini-hydro along with use of fuel cells and micro turbines will gain considerable momentum in the near future. A microgrid consists of clusters of load and distributed generators that operate as a single controllable system. The interconnection of the DG to the utility/grid through power electronic converters has raised concern about safe operation and protection of the equipments. Many innovative control techniques have been used for enhancing the stability of microgrid as for proper load sharing. The most common method is the use of droop characteristics for decentralized load sharing. Parallel converters have been controlled to deliver desired real power (and reactive power) to the system. Local signals are used as feedback to control converters, since in a real system, the distance between the converters may make the inter-communication impractical. The real and reactive power sharing can be achieved by controlling two independent quantities, frequency and fundamental voltage magnitude. In this thesis, an angle droop controller is proposed to share power amongst converter interfaced DGs in a microgrid. As the angle of the output voltage can be changed instantaneously in a voltage source converter (VSC), controlling the angle to control the real power is always beneficial for quick attainment of steady state. Thus in converter based DGs, load sharing can be performed by drooping the converter output voltage magnitude and its angle instead of frequency. The angle control results in much lesser frequency variation compared to that with frequency droop. An enhanced frequency droop controller is proposed for better dynamic response and smooth transition between grid connected and islanded modes of operation. A modular controller structure with modified control loop is proposed for better load sharing between the parallel connected converters in a distributed generation system. Moreover, a method for smooth transition between grid connected and islanded modes is proposed. Power quality enhanced operation of a microgrid in presence of unbalanced and non-linear loads is also addressed in which the DGs act as compensators. The compensator can perform load balancing, harmonic compensation and reactive power control while supplying real power to the grid A frequency and voltage isolation technique between microgrid and utility is proposed by using a back-to-back converter. As utility and microgrid are totally isolated, the voltage or frequency fluctuations in the utility side do not affect the microgrid loads and vice versa. Another advantage of this scheme is that a bidirectional regulated power flow can be achieved by the back-to-back converter structure. For accurate load sharing, the droop gains have to be high, which has the potential of making the system unstable. Therefore the choice of droop gains is often a tradeoff between power sharing and stability. To improve this situation, a supplementary droop controller is proposed. A small signal model of the system is developed, based on which the parameters of the supplementary controller are designed. Two methods are proposed for load sharing in an autonomous microgrid in rural network with high R/X ratio lines. The first method proposes power sharing without any communication between the DGs. The feedback quantities and the gain matrixes are transformed with a transformation matrix based on the line R/X ratio. The second method involves minimal communication among the DGs. The converter output voltage angle reference is modified based on the active and reactive power flow in the line connected at point of common coupling (PCC). It is shown that a more economical and proper power sharing solution is possible with the web based communication of the power flow quantities. All the proposed methods are verified through PSCAD simulations. The converters are modeled with IGBT switches and anti parallel diodes with associated snubber circuits. All the rotating machines are modeled in detail including their dynamics.

User facilitated congestion and attack mitigation

The IEEE Wireless LAN standard has been a true success story by enabling convenient, efficient and low-cost access to broadband networks for both private and professional use. However, the increasing density and uncoordinated operation of wireless access points, combined with constantly growing traffic demands have started hurting the users' quality of experience. On the other hand, the emerging ubiquity of wireless access has placed it at the center of attention for network attacks, which not only raises users' concerns on security but also indirectly affects connection quality due to proactive measures against security attacks. In this work, we introduce an integrated solution to congestion avoidance and attack mitigation problems through cooperation among wireless access points. The proposed solution implements a Partially Observable Markov Decision Process (POMDP) as an intelligent distributed control system. By successfully differentiating resource hampering attacks from overload cases, the control system takes an appropriate action in each detected anomaly case without disturbing the quality of service for end users. The proposed solution is fully implemented on a small-scale testbed, on which we present our observations and demonstrate the effectiveness of the system to detect and alleviate both attack and congestion situations.

Performance-driven live migration of multiple virtual machines in datacenters

Live migration of multiple Virtual Machines (VMs) has become an indispensible management activity in datacenters for application performance, load balancing, server consolidation. While state-of-the-art live VM migration strategies focus on the improvement of the migration performance of a single VM, little attention has been given to the case of multiple VMs migration. Moreover, existing works on live VM migration ignore the inter-VM dependencies, and underlying network topology and its bandwidth. Different sequences of migration and different allocations of bandwidth result in different total migration times and total migration downtimes. This paper concentrates on developing a multiple VMs migration scheduling algorithm such that the performance of migration is maximized. We evaluate our proposed algorithm through simulation. The simulation results show that our proposed algorithm can migrate multiple VMs on any datacenter with minimum total migration time and total migration downtime.

A random key genetic algorithm for live migration of multiple virtual machines in data centers

Live migration of multiple Virtual Machines (VMs) has become an integral management activity in data centers for power saving, load balancing and system maintenance. While state-of-the-art live migration techniques focus on the improvement of migration performance of an independent single VM, only a little has been investigated to the case of live migration of multiple interacting VMs. Live migration is mostly influenced by the network bandwidth and arbitrarily migrating a VM which has data inter-dependencies with other VMs may increase the bandwidth consumption and adversely affect the performances of subsequent migrations. In this paper, we propose a Random Key Genetic Algorithm (RKGA) that efficiently schedules the migration of a given set of VMs accounting both inter-VM dependency and data center communication network. The experimental results show that the RKGA can schedule the migration of multiple VMs with significantly shorter total migration time and total downtime compared to a heuristic algorithm.

Application of simulated annealing to data distribution for all-to-all comparison problems in homogeneous systems

Distributed systems are widely used for solving large-scale and data-intensive computing problems, including all-to-all comparison (ATAC) problems. However, when used for ATAC problems, existing computational frameworks such as Hadoop focus on load balancing for allocating comparison tasks, without careful consideration of data distribution and storage usage. While Hadoop-based solutions provide users with simplicity of implementation, their inherent MapReduce computing pattern does not match the ATAC pattern. This leads to load imbalances and poor data locality when Hadoop's data distribution strategy is used for ATAC problems. Here we present a data distribution strategy which considers data locality, load balancing and storage savings for ATAC computing problems in homogeneous distributed systems. A simulated annealing algorithm is developed for data distribution and task scheduling. Experimental results show a significant performance improvement for our approach over Hadoop-based solutions.

Distributed computing of all-to-all comparison problems in heterogeneous systems

The requirement of distributed computing of all-to-all comparison (ATAC) problems in heterogeneous systems is increasingly important in various domains. Though Hadoop-based solutions are widely used, they are inefficient for the ATAC pattern, which is fundamentally different from the MapReduce pattern for which Hadoop is designed. They exhibit poor data locality and unbalanced allocation of comparison tasks, particularly in heterogeneous systems. The results in massive data movement at runtime and ineffective utilization of computing resources, affecting the overall computing performance significantly. To address these problems, a scalable and efficient data and task distribution strategy is presented in this paper for processing large-scale ATAC problems in heterogeneous systems. It not only saves storage space but also achieves load balancing and good data locality for all comparison tasks. Experiments of bioinformatics examples show that about 89\% of the ideal performance capacity of the multiple machines have be achieved through using the approach presented in this paper.

Data distribution and task scheduling for distributed computing of all-to-all comparison problems

This research studied distributed computing of all-to-all comparison problems with big data sets. The thesis formalised the problem, and developed a high-performance and scalable computing framework with a programming model, data distribution strategies and task scheduling policies to solve the problem. The study considered storage usage, data locality and load balancing for performance improvement in solving the problem. The research outcomes can be applied in bioinformatics, biometrics and data mining and other domains in which all-to-all comparisons are a typical computing pattern.

Rail power conditioner technology in Australian Heavy Haul Railway: A case study

Aurizon, Australia's largest freight railway operator, is investigating the use of Rail Power Conditioner (RPC) technology for load balancing, reactive power compensation and harmonic filtering. The new technology has the capability of replacing Static VAr Compensators (SVC) and Harmonic Filters, and is expected to have a significant impact on the overall costs of railway electrification. This paper presents the theoretical analysis of the real and reactive power flows in an RPC used to balance active powers in an existing V/V feeder station. This informed an RPC feasibility study undertaken at four existing Aurizon's feeder stations with V/V connected transformers.

Data-aware task scheduling for all-to-all comparison problems in heterogeneous distributed systems

Solving large-scale all-to-all comparison problems using distributed computing is increasingly significant for various applications. Previous efforts to implement distributed all-to-all comparison frameworks have treated the two phases of data distribution and comparison task scheduling separately. This leads to high storage demands as well as poor data locality for the comparison tasks, thus creating a need to redistribute the data at runtime. Furthermore, most previous methods have been developed for homogeneous computing environments, so their overall performance is degraded even further when they are used in heterogeneous distributed systems. To tackle these challenges, this paper presents a data-aware task scheduling approach for solving all-to-all comparison problems in heterogeneous distributed systems. The approach formulates the requirements for data distribution and comparison task scheduling simultaneously as a constrained optimization problem. Then, metaheuristic data pre-scheduling and dynamic task scheduling strategies are developed along with an algorithmic implementation to solve the problem. The approach provides perfect data locality for all comparison tasks, avoiding rearrangement of data at runtime. It achieves load balancing among heterogeneous computing nodes, thus enhancing the overall computation time. It also reduces data storage requirements across the network. The effectiveness of the approach is demonstrated through experimental studies.