A study of adaptive load balancing algorithms for distributed systems

With the advent of distributed computer systems with a largely transparent user interface, new questions have arisen regarding the management of such an environment by an operating system. One fertile area of research is that of load balancing, which attempts to improve system performance by redistributing the workload submitted to the system by the users. Early work in this field concentrated on static placement of computational objects to improve performance, given prior knowledge of process behaviour. More recently this has evolved into studying dynamic load balancing with process migration, thus allowing the system to adapt to varying loads. In this thesis, we describe a simulated system which facilitates experimentation with various load balancing algorithms. The system runs under UNIX and provides functions for user processes to communicate through software ports; processes reside on simulated homogeneous processors, connected by a user-specified topology, and a mechanism is included to allow migration of a process from one processor to another. We present the results of a study of adaptive load balancing algorithms, conducted using the aforementioned simulated system, under varying conditions; these results show the relative merits of different approaches to the load balancing problem, and we analyse the trade-offs between them. Following from this study, we present further novel modifications to suggested algorithms, and show their effects on system performance.

The relative performance of scalable load balancing algorithms in loosely-coupled distributed systems. Available in 2 volumes.

The computer systems of today are characterised by data and program control that are distributed functionally and geographically across a network. A major issue of concern in this environment is the operating system activity of resource management for different processors in the network. To ensure equity in load distribution and improved system performance, load balancing is often undertaken. The research conducted in this field so far, has been primarily concerned with a small set of algorithms operating on tightly-coupled distributed systems. More recent studies have investigated the performance of such algorithms in loosely-coupled architectures but using a small set of processors. This thesis describes a simulation model developed to study the behaviour and general performance characteristics of a range of dynamic load balancing algorithms. Further, the scalability of these algorithms are discussed and a range of regionalised load balancing algorithms developed. In particular, we examine the impact of network diameter and delay on the performance of such algorithms across a range of system workloads. The results produced seem to suggest that the performance of simple dynamic policies are scalable but lack the load stability of more complex global average algorithms.

Management of services differentiation and guarantee in IEEE 802.11e wireless LANs

In this paper, we study the management and control of service differentiation and guarantee based on enhanced distributed function coordination (EDCF) in IEEE 802.11e wireless LANs. Backoff-based priority schemes are the major mechanism for Quality of Service (QoS) provisioning in EDCF. However, control and management of the backoff-based priority scheme are still challenging problems. We have analysed the impacts of backoff and Inter-frame Space (IFS) parameters of EDCF on saturation throughput and service differentiation. A centralised QoS management and control scheme is proposed. The configuration of backoff parameters and admission control are studied in the management scheme. The special role of access point (AP) and the impact of traffic load are also considered in the scheme. The backoff parameters are adaptively re-configured to increase the levels of bandwidth guarantee and fairness on sharing bandwidth. The proposed management scheme is evaluated by OPNET. Simulation results show the effectiveness of the analytical model based admission control scheme. ©2005 IEEE.

Performance management of event processing systems

This thesis is a study of performance management of Complex Event Processing (CEP) systems. Since CEP systems have distinct characteristics from other well-studied computer systems such as batch and online transaction processing systems and database-centric applications, these characteristics introduce new challenges and opportunities to the performance management for CEP systems. Methodologies used in benchmarking CEP systems in many performance studies focus on scaling the load injection, but not considering the impact of the functional capabilities of CEP systems. This thesis proposes the approach of evaluating the performance of CEP engines’ functional behaviours on events and develops a benchmark platform for CEP systems: CEPBen. The CEPBen benchmark platform is developed to explore the fundamental functional performance of event processing systems: filtering, transformation and event pattern detection. It is also designed to provide a flexible environment for exploring new metrics and influential factors for CEP systems and evaluating the performance of CEP systems. Studies on factors and new metrics are carried out using the CEPBen benchmark platform on Esper. Different measurement points of response time in performance management of CEP systems are discussed and response time of targeted event is proposed to be used as a metric for quality of service evaluation combining with the traditional response time in CEP systems. Maximum query load as a capacity indicator regarding to the complexity of queries and number of live objects in memory as a performance indicator regarding to the memory management are proposed in performance management of CEP systems. Query depth is studied as a performance factor that influences CEP system performance.

Using cognitive load theory to interpret student difficulties with a problem-based learning approach to engineering education:a case study

This article reports on an investigationwith first year undergraduate ProductDesign and Management students within a School of Engineering and Applied Science. The students at the time of this investigation had studied fundamental engineering science and mathematics for one semester. The students were given an open ended, ill-formed problem which involved designing a simple bridge to cross a river.They were given a talk on problemsolving and given a rubric to follow, if they chose to do so.They were not given any formulae or procedures needed in order to resolve the problem. In theory, they possessed the knowledge to ask the right questions in order tomake assumptions but, in practice, it turned out they were unable to link their a priori knowledge to resolve this problem. They were able to solve simple beam problems when given closed questions. The results show they were unable to visualize a simple bridge as an augmented beam problem and ask pertinent questions and hence formulate appropriate assumptions in order to offer resolutions.

An improved memory management scheme for large scale graph computing engine GraphChi

GraphChi is the first reported disk-based graph engine that can handle billion-scale graphs on a single PC efficiently. GraphChi is able to execute several advanced data mining, graph mining and machine learning algorithms on very large graphs. With the novel technique of parallel sliding windows (PSW) to load subgraph from disk to memory for vertices and edges updating, it can achieve data processing performance close to and even better than those of mainstream distributed graph engines. GraphChi mentioned that its memory is not effectively utilized with large dataset, which leads to suboptimal computation performances. In this paper we are motivated by the concepts of 'pin ' from TurboGraph and 'ghost' from GraphLab to propose a new memory utilization mode for GraphChi, which is called Part-in-memory mode, to improve the GraphChi algorithm performance. The main idea is to pin a fixed part of data inside the memory during the whole computing process. Part-in-memory mode is successfully implemented with only about 40 additional lines of code to the original GraphChi engine. Extensive experiments are performed with large real datasets (including Twitter graph with 1.4 billion edges). The preliminary results show that Part-in-memory mode memory management approach effectively reduces the GraphChi running time by up to 60% in PageRank algorithm. Interestingly it is found that a larger portion of data pinned in memory does not always lead to better performance in the case that the whole dataset cannot be fitted in memory. There exists an optimal portion of data which should be kept in the memory to achieve the best computational performance.

How does problem based learning fit with cognitive load theory?

This paper reports on an investigation with first year undergraduate Product Design and Management students within a School of Engineering. The students at the time of this investigation had studied fundamental engineering science and mathematics for one semester. The students were given an open ended, ill formed problem which involved designing a simple bridge to cross a river. They were given a talk on problem solving and given a rubric to follow, if they chose to do so. They were not given any formulae or procedures needed in order to resolve the problem. In theory, they possessed the knowledge to ask the right questions in order to make assumptions but, in practice, it turned out they were unable to link their a priori knowledge to resolve this problem. They were able to solve simple beam problems when given closed questions. The results show they were unable to visualise a simple bridge as an augmented beam problem and ask pertinent questions and hence formulate appropriate assumptions in order to offer resolutions.

QoE-aware power management in vehicle-to-grid networks:a matching-theoretic approach

Frequency, time and places of charging and discharging have critical impact on the Quality of Experience (QoE) of using Electric Vehicles (EVs). EV charging and discharging scheduling schemes should consider both the QoE of using EV and the load capacity of the power grid. In this paper, we design a traveling plan-aware scheduling scheme for EV charging in driving pattern and a cooperative EV charging and discharging scheme in parking pattern to improve the QoE of using EV and enhance the reliability of the power grid. For traveling planaware scheduling, the assignment of EVs to Charging Stations (CSs) is modeled as a many-to-one matching game and the Stable Matching Algorithm (SMA) is proposed. For cooperative EV charging and discharging in parking pattern, the electricity exchange between charging EVs and discharging EVs in the same parking lot is formulated as a many-to-many matching model with ties, and we develop the Pareto Optimal Matching Algorithm (POMA). Simulation results indicates that the SMA can significantly improve the average system utility for EV charging in driving pattern, and the POMA can increase the amount of electricity offloaded from the grid which is helpful to enhance the reliability of the power grid.