Artificial life techniques for load balancing in computational grids

Load balancing is a very important and complex problem in computational grids. A computational grid differs from traditional high performance computing systems in the heterogeneity of the computing nodes and communication links, as well as background workloads that may be present in the computing nodes. There is a need to develop algorithms that could capture this complexity yet can be easily implemented and used to solve a wide range of load balancing scenarios. Artificial life techniques have been used to solve a wide range of complex problems in recent times. The power of these techniques stems from their capability in searching large search spaces, which arise in many combinatorial optimization problems, very efficiently. This paper studies several well-known artificial life techniques to gauge their suitability for solving grid load balancing problems. Due to their popularity and robustness, a genetic algorithm (GA) and tabu search (TS) are used to solve the grid load balancing problem. The effectiveness of each algorithm is shown for a number of test problems, especially when prediction information is not fully accurate. Performance comparisons with Min-min, Max-min, and Sufferage are also discussed.     

A bit-masking oriented data structure for evolutionary operators implementation in genetic algorithms

 In the present paper a special bit-masking oriented data structure for an improved implementation of crossover and mutation operators in genetic algorithms is shown. The developed data structure performs evolutionary operators in two separate steps: crossover and mutation mask fill and a special boolean based function application. Both phases are optimized to reach a more efficient, fast and flexible genetic reproduction than standard implementations. The method has been powered adding a multi-layered, bit-masking oriented data structure and a boolean operation based control mixer, allowing special blended crossover operators obtained by superposition of the standard ones. Several examples of crossover schemes produced by these extended controls are presented. In addition, a special purpose crossover scheme, capable to process at the same time two distinct groups of design variables with separate crossover schemes is shown, in order to improve efficiency and convergence speed of some discrete/continuous optimization problems. Finally, to highlight further capabilities of the bit-masking approach, a special single-step version of an evolutionary direction operator is also illustrated.     

Data partitioning and load balancing in parallel disk systems

Parallel disk systems provide opportunities for exploiting I/O parallelism in two possible ways, namely via inter-request and intra-request parallelism. In this paper, we discuss the main issues in performance tuning of such systems, namely striping and load balancing, and show their relationship to response time and throughput. We outline the main components of an intelligent, self-reliant file system that aims to optimize striping by taking into account the requirements of the applications, and performs load balancing by judicious file allocation and dynamic redistributions of the data when access patterns change. Our system uses simple but effective heuristics that incur only little overhead. We present performance experiments based on synthetic workloads and real-life traces. Received May 17, 1994 / Accepted June 9, 1997     

Application of computational intelligence techniques in active networks

 Computational intelligence techniques have been successfully used for solving control problems in packet-switching network architectures. The introduction of active networking adds a high degree of flexibility in customizing the network infrastructure and introducing new functionality. Therefore, there is a clear need for investigating both the applicability of computational intelligence techniques in this new networking environment, as well as the provisions of active networking technology that computational intelligence techniques can exploit for improved operation. We report on the characteristics of these technologies, their synergy and on outline recent efforts in the design of a computational intelligence toolkit and its application to routing on a novel active networking environment.     

An application of genetic and evolutive algorithms to unit commitment problem

 Here is introduced an application of the Genetic and Evolutive Algorithms to the Unit Commitment Problem. It is a mixed integer problem of constrained non linear combinatorial optimization. The many constraints make the problem very complex. Three cases of study on the problem have been faced, characterized by crescent grades of completeness/ difficulties in order to understand which are the advantages and the difficulties which arise from the evolutive approach. In the cases of study have been faced dimensions of the problem significant in practice: from 10 up to 1000 generators.     

Fuzzy evolutionary computing

 This paper elaborates on a new paradigm of computing embracing fuzzy sets and evolutionary methods (specially genetic algorithms). We discuss conceptual and algorithmic enhancements to the individual methods. Fuzzy sets are geared toward granular information processing. Evolutionary computing are population-based optimization methods. In this way, as being components of any hybrid structure, they naturally complement each other. The study reveals a number of representative symbiotic links between fuzzy and genetic computing and provides with relevant illustrative examples.     

A new evolutionary algorithm combining simulated annealing and genetic programming for relevance feedback in fuzzy information retrieval systems

 Relevance feedback techniques have demonstrated to be a powerful means to improve the results obtained when a user submits a query to an information retrieval system as the world wide web search engines. These kinds of techniques modify the user original query taking into account the relevance judgements provided by him on the retrieved documents, making it more similar to those he judged as relevant. This way, the new generated query permits to get new relevant documents thus improving the retrieval process by increasing recall. However, although powerful relevance feedback techniques have been developed for the vector space information retrieval model and some of them have been translated to the classical Boolean model, there is a lack of these tools in more advanced and powerful information retrieval models such as the fuzzy one. In this contribution we introduce a relevance feedback process for extended Boolean (fuzzy) information retrieval systems based on a hybrid evolutionary algorithm combining simulated annealing and genetic programming components. The performance of the proposed technique will be compared with the only previous existing approach to perform this task, Kraft et al.'s method, showing how our proposal outperforms the latter in terms of accuracy and sometimes also in time consumption. Moreover, it will be showed how the adaptation of the retrieval threshold by the relevance feedback mechanism allows the system effectiveness to be increased.     

Mining linguistic browsing patterns in the world wide web

 World-wide-web applications have grown very rapidly and have made a significant impact on computer systems. Among them, web browsing for useful information may be most commonly seen. Due to its tremendous amounts of use, efficient and effective web retrieval has thus become a very important research topic in this field. Data mining is the process of extracting desirable knowledge or interesting patterns from existing databases for a certain purpose. In this paper, we use the data mining techniques to discover relevant browsing behavior from log data in web servers, thus being able to help make rules for retrieval of web pages. The browsing time of a customer on each web page is used to analyze the retrieval behavior. Since the data collected are numeric, fuzzy concepts are used to process them and to form linguistic terms. A sophisticated web-mining algorithm is thus proposed to find relevant browsing behavior from the linguistic data. Each page uses only the linguistic term with the maximum cardinality in later mining processes, thus making the number of fuzzy regions to be processed the same as the number of the pages. Computational time can thus be greatly reduced. The patterns mined out thus exhibit the browsing behavior and can be used to provide some appropriate suggestions to web-server managers.     

进化算法的计算结果呈现方式

进化算法（Evolutionary Algorithms,EAs）作为求解非线性规划问题的有效求解工具已经越来越受到工程和优化领域的国内外专家和学者的重视,进化算法类的文章在世界上各种期刊中占据了大量比例。目前仍有很多刚刚从事进化算法理论与实践方面研究的国内学者对如何表现进化算法的计算结果比较迷茫。为此对于算法的计算结果展现方面进行了阐述。     

A new mutation operator for evolutionary airfoil design

A new mutation operator, ℳ _ijn , capable of operating on a set of adjacent bits in one single step, is introduced. Its features are examined and compared against those of the classical bit–flip mutation. A simple Evolutionary Algorithm, ℳ–EA, based only on selection and ℳ _ijn , is described. This algorithm is used for the solution of an industrial problem, the Inverse Airfoil Design optimization, characterized by high search time to achieve satisfying solutions, and its performance is compared against that offered by a classical binary Genetic Algorithm. The experiments show for our algorithm a noticeable reduction in the time needed to reach a solution of acceptable quality, thus they prove the effectiveness of the proposed operator and its superiority to GAs for the problem at hand.     

Analysis of navigation behaviour in web sites integrating multiple information systems

Abstract. The analysis of web usage has mostly focused on sites composed of conventional static pages. However, huge amounts of information available in the web come from databases or other data collections and are presented to the users in the form of dynamically generated pages. The query interfaces of such sites allow the specification of many search criteria. Their generated results support navigation to pages of results combining cross-linked data from many sources. For the analysis of visitor navigation behaviour in such web sites, we propose the web usage miner (WUM), which discovers navigation patterns subject to advanced statistical and structural constraints. Since our objective is the discovery of interesting navigation patterns, we do not focus on accesses to individual pages. Instead, we construct conceptual hierarchies that reflect the query capabilities used in the production of those pages. Our experiments with a real web site that integrates data from multiple databases, the German SchulWeb, demonstrate the appropriateness of WUM in discovering navigation patterns and show how those discoveries can help in assessing and improving the quality of the site. Received June 21, 1999 / Accepted December 24, 1999     

Learning classifier systems from a reinforcement learning perspective

 We analyze learning classifier systems in the light of tabular reinforcement learning. We note that although genetic algorithms are the most distinctive feature of learning classifier systems, it is not clear whether genetic algorithms are important to learning classifiers systems. In fact, there are models which are strongly based on evolutionary computation (e.g., Wilson's XCS) and others which do not exploit evolutionary computation at all (e.g., Stolzmann's ACS). To find some clarifications, we try to develop learning classifier systems “from scratch”, i.e., starting from one of the most known reinforcement learning technique, Q-learning. We first consider thebasics of reinforcement learning: a problem modeled as a Markov decision process and tabular Q-learning. We introduce a formal framework to define a general purpose rule-based representation which we use to implement tabular Q-learning. We formally define generalization within rules and discuss the possible approaches to extend our rule-based Q-learning with generalization capabilities. We suggest that genetic algorithms are probably the most general approach for adding generalization although they might be not the only solution.     

Power optimization for dynamic configuration in heterogeneous web server clusters

To reduce the environmental impact, it is essential to make data centers green, by turning off servers and tuning their speeds for the instantaneous load offered, that is, determining the dynamic configuration in web server clusters. We model the problem of selecting the servers that will be on and finding their speeds through mixed integer programming; we also show how to combine such solutions with control theory. For proof of concept, we implemented this dynamic configuration scheme in a web server cluster running Linux, with soft real-time requirements and QoS control, in order to guarantee both energy-efficiency and good user experience. In this paper, we show the performance of our scheme compared to other schemes, a comparison of a centralized and a distributed approach for QoS control, and a comparison of schemes for choosing speeds of servers.     

Solving fuzzy equations using evolutionary algorithms and neural nets

 In this paper we use evolutionary algorithms and neural nets to solve fuzzy equations. In Part I we: (1) first introduce our three solution methods for solving the fuzzy linear equation AˉXˉ + Bˉ= Cˉ; for Xˉ and (2) then survey the results for the fuzzy quadratic equations, fuzzy differential equations, fuzzy difference equations, fuzzy partial differential equations, systems of fuzzy linear equations, and fuzzy integral equations; and (3) apply an evolutionary algorithm to construct one of the solution types for the fuzzy eigenvalue problem. In Part II we: (1) first discuss how to design and train a neural net to solve AˉXˉ + Bˉ= Cˉ for Xˉ and (2) then survey the results for systems of fuzzy linear equations and the fuzzy quadratic.     

Adaptive and scalable load balancing for metadata server cluster in cloud-scale file systems

Big data is an emerging term in the storage industry, and it is data analytics on big storage, i.e., Cloud-scale storage. In Cloud-scale (or EB-scale) file systems, load balancing in request workloads across a metadata server cluster is critical for avoiding performance bottlenecks and improving quality of services.Many good approaches have been proposed for load balancing in distributed file systems. Some of them pay attention to global namespace balancing, making metadata distribution across metadata servers as uniform as possible. However, they do not work well in skew request distributions, which impair load balancing but simultaneously increase the effectiveness of caching and replication. In this paper, we propose Cloud Cache (C²), an adaptive and scalable load balancing scheme for metadata server cluster in EB-scale file systems. It combines adaptive cache diffusion and replication scheme to cope with the request load balancing problem, and it can be integrated into existing distributed metadata management approaches to efficiently improve their load balancing performance. C² runs as follows: 1) to run adaptive cache diffusion first, if a node is overloaded, loadshedding will be used; otherwise, load-stealing will be used; and 2) to run adaptive replication scheme second, if there is a very popular metadata item (or at least two items) causing a node be overloaded, adaptive replication scheme will be used, in which the very popular item is not split into several nodes using adaptive cache diffusion because of its knapsack property. By conducting performance evaluation in trace-driven simulations, experimental results demonstrate the efficiency and scalability of C².     

Adaptive mutation in genetic algorithms

 In Genetic Algorithms mutation probability is usually assigned a constant value, therefore all chromosome have the same likelihood of mutation irrespective of their fitness. It is shown in this paper that making mutation a function of fitness produces a more efficient search. This function is such that the least significant bits are more likely to be mutated in high-fitness chromosomes, thus improving their accuracy, whereas low-fitness chromosomes have an increased probability of mutation, enhancing their role in the search. In this way, the chance of disrupting a high-fitness chromosome is decreased and the exploratory role of low-fitness chromosomes is best exploited. The implications of this new mutation scheme are assessed with the aid of numerical examples.     

A genetic-based methodology for hybrid electric vehicles sizing

 As private transport concerns, the global challenge of this millennium is the reduction of carbon dioxide emissions from passenger cars by improving fuel economy without sacrificing the vehicle performance. Hybrid electric vehicles powertrain, combining electric motor with an auxiliary power unit, can improve effectively the vehicle performance and fuel economy, reducing at the same time the effects of the use of private cars on the air quality of the cities. These advantages can be achieved only if the design of the powertrain is inspired to the minimisation of the main figures of merit holding in consideration many general aspects and variables. As supporting methodology in developing this difficult activity, a genetic-based sizing methodology will be presented. It will be aimed to minimise a function objective which takes into account not only technical specifications but also environmental, social, and economic aspects. Some interesting simulation results will be reported to prove the validity of the methodology, which will contribute to a substantial reduction of the pollutant emissions from hybrid electric vehicles.     

Structured population-based incremental learning

 We investigate a recently developed abstraction of genetic algorithms (GAs) in which a population of GAs in any generation is represented by a single vector whose elements are the probabilities of the corresponding bit positions being equivalent to 1. The process of evolution is represented by learning the elements of the probability vector; the method is clearly linked to the artificial neural network (ANN) method of competitive learning. We use techniques from ANNs to extend the applicability of the method to non-static problems, to multi-objective criteria, to multi-modal problems and to creating an order on a set of sub-populations.     

Granular neural web agents for stock prediction

 A granular neural Web-based stock prediction agent is developed using the granular neural network (GNN) that can discover fuzzy rules. Stock data sets are downloaded from www.yahoo.com website. These data sets are inserted into the database tables using a java program. Then, the GNN is trained using sample data for any stock. After learning from the past stock data, the GNN is able to use discover fuzzy rules to make future predictions. After doing simulations with six different stocks (msft, orcl, dow, csco, ibm, km), it is conclusive that the granular neural stock prediction agent is giving less average errors with large amount of past training data and high average errors in case of fewer amounts of past training data. Java Servlets, Java Script and jdbc are used. SQL is used as the back-end database. The performance of the GNN algorithm is compared with the performance of the BP algorithm by training the same set of data and predicting the future stock values. The average error of the GNN is less than that of BP algorithm.