Ahmet Cosar

Genetic Algorithm for the Multiple-Query Optimization Problem

Producing answers to a set of queries with common tasks efficiently is known as the multiple-query optimization (MQO) problem. Each query can have several alternative evaluation plans, each with a different set of tasks. Therefore, the goal of MQO is to choose the right set of plans for queries which minimizes the total execution time by performing common tasks only once. Since MQO is an NP-hard problem, several, mostly heuristics based, solutions have been proposed for solving it. To the best of our knowledge, this correspondence is the first attempt to solve MQO using an evolutionary technique, genetic algorithms

An Evolutionary Genetic Algorithm for Optimization of Distributed Database Queries

High performance low cost PC hardware, and high speed LAN/WAN technologies make distributed database(DDB) systems an attractive research area. Since Dynamic programming is not feasible for optimizing queries in a DDB, we propose a GA based query optimizer and compare its performance to random and optimal algorithms. We analyzed a set of possible GA parameters and determined that two-point truncate technique using GA gives the best results. New mutation and crossover operators have also been defined and experimentally analyzed. We performed experiments on a synthetic database with replicated relations, but no horizontal or vertical fragmentation. Network links are assumed to be gigabit Ethernet. Comparisons with optimal results found by exhaustive search show that our new GA formulation performs only 20% off the optimal results and we have achieved a 50% improvement over a previous GA based algorithm.

A robust Island Parallel Genetic Algorithm for the Quadratic Assignment Problem

The Quadratic Assignment Problem (QAP) is a difficult and important problem studied in the domain of combinatorial optimisation. It is possible to solve QAP instances with 10--20 facilities using exhaustive parallel algorithms within a few days on a cluster machine. However, large QAP instances with more than 100 facilities are not solvable using exhaustive techniques. We have explored a variety of Genetic Algorithm crossover operators for this problem and verified its performance experimentally using well-known instances from the QAPLIB library. By increasing the number of processors, generations and population sizes we have been able to find solutions that are the same as (or very close to) the best reported solutions for large QAP instances in QAPLIB. In order to parallelise the Genetic Algorithm we generate and evolve separate solution pools on each cluster processor, using an island model. This model exchanges 10% of each processor’s solutions at the initial stages of optimisation. We show experimentally that both execution times and solution qualities are improved for large QAP instances by using our Island Parallel Genetic Algorithm.

A novel multistart hyper-heuristic algorithm on the grid for the quadratic assignment problem

Hyper-heuristics introduce novel approaches for solving challenging combinatorial optimization problems by operating over a set of low level (meta)-heuristics. This is achieved by an evolutionary selection mechanism that controls and combines the strengths of the low level (meta)-heuristics. In this study, we propose a high-performance MultiStart Hyper-heuristic algorithm (MSH-QAP) on the grid for the solution of the Quadratic Assignment Problem (QAP). MSH-QAP algorithm makes use of state-of-the-art (meta)-heuristics, Simulated Annealing (SA), Robust Tabu Search (RTS), Ant Colony Optimization (FAnt), and Breakout Local Search (BLS) that have been reported among the best performing algorithms for the solution of difficult QAP instances in standard benchmark libraries. In the first phase of the algorithm, the most appropriate (meta)-heuristic with its near-optimal parameter settings is selected by using a genetic algorithm optimization layer that uses a self-adaptive parameter setting method for the given problem instance. In the second phase, if an optimal solution cannot be found, selected best performing (meta)-heuristic (with its finely adjusted parameter settings) is executed on the grid using parallel processing and performing several multistarts in order to increase the quality of the discovered solution. MSH-QAP algorithm is tested on 134 problem instances of the QAPLIB benchmark and is shown to be able to solve 122 of the instances exactly. The overall deviation for the problem instances is obtained as 0.013% on the average.

Optimization of one-dimensional Bin Packing Problem with island parallel grouping genetic algorithms

Abstract The well-known one-dimensional Bin Packing Problem (BPP) of whose variants arise in many real life situations is a challenging NP-Hard combinatorial optimization problem. Metaheuristics are widely used optimization tools to find (near-) optimal solutions for solving large problem instances of BPP in reasonable running times. With this study, we propose a set of robust and scalable hybrid parallel algorithms that take advantage of parallel computation techniques, evolutionary grouping genetic metaheuristics, and bin-oriented heuristics to obtain solutions for large scale one-dimensional BPP instances. A total number of 1318 benchmark problems are examined with the proposed algorithms and it is shown that optimal solutions for 88.5% of these instances can be obtained with practical optimization times while solving the rest of the problems with no more than one extra bin. When the results are compared with the existing state-of-the-art heuristics, the developed parallel hybrid grouping genetic algorithms can be considered as one of the best one-dimensional BPP algorithms in terms of computation time and solution quality.

Discovering better navigation sequences for the session construction problem

In this paper, we propose a novel page view based session model and session construction method to address the Web Usage Mining (WUM) problem. Unlike the simple session models, where sessions are sequences of web pages requested from the server (or served from a browser/proxy cache) and viewed in the browser (which may not guarantee a direct relationship between subsequent web pages in the session), we define a more realistic session model in which a session is a set of paths traversed in the web graph that corresponds to a user navigation performed by following links on web pages. We define the session construction process from raw server logs as a new graph problem and present a novel algorithm, Smart-SRA (Smart Session Reconstruction Algorithm), to solve this problem efficiently. An experimental evaluation based on data collected from real web access scenarios showed that Smart-SRA produces more accurate user sessions than the session construction methods found in the literature.

Dynamic programming solution for multiple query optimization problem

erved The “multiple query optimization” (MQO) prob lem has been studied in the database literature s 1980s. MQO tries to reduce the execution cost o group of queries by performing common tasks o once, whereas traditional query optimization cons ers a single query at a time. MQO has been formula in [6] as an NP-complete optimization problem whe several heuristic functions are used to direct an A∗ search. Later in [7], and [2] a more informed cost e mation function, which is more expensive to calcula has been used to reduce the number of states exp by theA∗ search. Recently research emphasis has b on efficiently generating alternative plans that ma mize shared operations [4,5]. Roy et al. [4] descr a greedy heuristic for genera ting promising alternative

Multi-objective evolutionary routing protocol for efficient coverage in mobile sensor networks

Individual sensors in wireless mobile sensor networks (MSNs) can move in search of coverage region for the sensing accuracy and for reaching the most efficient topology. Besides, sensors’ clustering is crucial for achieving an efficient network performance. Although MSNs have been an area of many research efforts in recent years, integrating the coverage problem of MSNs with the efficient routing requirement that will maximize the network lifetime is still missing. In this paper, we consider the coverage optimization problem where the location of a given number of mobile sensors needs to be re-decided such that the sensed data from the detected targets can be routed more efficiently to the sink and thus increasing the network lifetime. We formulate this NP-complete problem as a multi-objective optimization (MOO) problem, with two conflicting and correlated objectives; aiming at high coverage as well as longevity of network lifetime. The Non-Dominated Sorting Genetic Algorithm-II (NSGA-II) is utilized as a cluster-based routing protocol to tackle this MOO problem. Each round of the proposed NSGA-II based routing protocol creates a set of near-Pareto-optimal solutions containing a number of non-dominated solutions, in which the sink can pick up and distribute the one with high coverage to form the clustered routes. Heuristic operators are also proposed to enhance the quality of the solutions. Simulation results are provided to illustrate the effectiveness and performance of the proposed evolutionary algorithm.

Improving the performance of Hadoop Hive by sharing scan and computation tasks

MapReduce is a popular programming model for executing time-consuming analytical queries as a batch of tasks on large scale data clusters. In environments where multiple queries with similar selection predicates, common tables, and join tasks arrive simultaneously, many opportunities can arise for sharing scan and/or join computation tasks. Executing common tasks only once can remarkably reduce the total execution time of a batch of queries. In this study, we propose a Multiple Query Optimization framework, SharedHive, to improve the overall performance of Hadoop Hive, an open source SQL-based data warehouse using MapReduce. SharedHive transforms a set of correlated HiveQL queries into a new set of insert queries that will produce all of the required outputs within a shorter execution time. It is experimentally shown that SharedHive achieves significant reductions in total execution times of TPC-H queries.

Performance Comparison of Pattern Discovery Methods on Web Log Data

One of the popular trends in computer science has been development of intelligent web-based systems. Demand for such systems forces designers to make use of knowledge discovery techniques on web server logs. Web usage mining has become a major area of knowledge discovery on World Wide Web. Frequent pattern discovery is one of the main issues in web usage mining. These frequent patterns constitute the basic information source for intelligent web-based systems. In this paper; frequent pattern mining algorithms for web log data and their performance comparisons are examined. Our study is mainly focused on finding suitable pattern mining algorithms for web server logs.