Journal of Advances in Computer Engineering and Technology
Volume:1 Issue: 4, Autumn 2015

 One of the important problems in grid environments is data replication in grid sites. Reliability and availability of data replication in some cases is considered low. To separate sites with high reliability and high availability of sites with low availability and low reliability, clustering can be used. In this study, the data grid dynamically evaluate and predict the condition of the sites. The reliability and availability of sites were calculated and it was used to make decisions to replicate data. With these calculations, we have information on the locations of users in grid with reliability and availability or cost commensurate with the value of the work they did. This information can be downloaded from users who are willing to send them data with suitable reliability and availability. Simulation results show that the addition of the two parameters, reliability and availability, assessment criteria have been improved in certain access patterns.

One of the most important and valuable goal of software development life cycle is software cost estimation or SCE. During the recent years, SCE has attracted the attention of researchers due to huge amount of software project requests. There have been proposed so many models using heuristic and meta-heuristic algorithms to do machine learning process for SCE. COCOMO81 is one of the most popular models for SCE proposed by Barry Boehm in 1981. However COCOMO81 is an old estimation model, it has been widely used for the purpose of cost estimation in its new forms. In this paper, the Imperialism Competition Algorithm (ICA) has been employed to tune the COCOMO81 parameters. Experimental results show that in the separated COCOMO81 dataset, ICA can estimate the COCOMO81 model parameters such that the performance parameters are significantly improved. The proposed hybrid model is flexible enough to tune the parameters for any data sets in form of COCOMO81.

Cloud computing refers to services that run in a distributed network and are accessible through common internet protocols. It merges a lot of physical resources and offers them to users as services according to service level agreement. Therefore, resource management alongside with task scheduling has direct influence on cloud networks’ performance and efficiency. Presenting a proper scheduling method can lead to efficiency of resources by decreasing response time and costs. This paper studies the existing approaches of task scheduling and resource allocation in cloud infrastructures and assessment of their advantages and disadvantages. Afterwards, a compound algorithm is presented in order to allocate tasks to resources properly and decrease runtime. The proposed algorithm is built according to conditions of compounding Min-min and Sufferage algorithms. In the proposed algorithm, task allocation between machines takes place alternatively and with continuous change of scheduling algorithms. The main idea of the proposed algorithm is to concentrate on the number of tasks instead of the existing resources. The simulation results reveal that the proposed algorithm can achieve higher performance in decreasing response time.

Data visualization is the process of transforming data, information, and knowledge into visual form, making use of humans’ natural visual capabilities which reveals relationships in data sets that are not evident from the raw data, by using mathematical techniques to reduce the number of dimensions in the data set while preserving the relevant inherent properties. In this paper, we formulated data visualization as a Quadric Assignment Problem (QAP), and then presented an Artificial Bee Colony (ABC) to solve the resulted discrete optimization problem. The idea behind this approach is to provide mechanisms based on ABC to overcome trapped in local minima and improving the resulted solutions. To demonstrate the application of ABC on discrete optimization in data visualization, we used a database of electricity load and compared the results to other popular methods such as SOM, MDS and Sammon's map. The results show that QAP-ABC has high performance with compared others.

Todays, in many modern applications, we search for frequent and repeating patterns in the analyzed data sets. In this search, we look for patterns that frequently appear in data set and mark them as frequent patterns to enable users to make decisions based on these discoveries. Most algorithms presented in the context of data stream mining and frequent pattern detection, work either on uncertain data, or use the sliding window model to assess data streams. Sliding window model uses a fixed-size window to only maintain the most recently inserted data and ignores all previous data (or those that are out of its window). Many real-world applications however require maintaining all inserted or obtained data. Therefore, the question arises that whether other window models can be used to find frequent patterns in dynamic streams of uncertain data.In this paper, we used landmark window model and time-fading model to answer that question. The method presented in the form of proposed algorithm, which uses the idea of landmark window model to find frequent patterns in the relational and uncertain data streams, shows a better performance in finding functional dependencies than other methods in this field. Another advantage of this method compared with other methods is that it shows tuples that do not follow a single dependency. This feature can be used to detect inconsistent data in a data set.

 In this paper we have tried to develop an altered version of the artificial bee colony algorithm which is inspired from and combined with the meta-heuristic algorithm of firefly. In this method, we have tried to change the main equation of searching within the original ABC algorithm. On this basis, a new combined equation was used for steps of employed bees and onlooker bees. For this purpose, we had to define several new parameters for improving the quality of the proposed method. In this regard, we have introduced two new parameters to the method. The new method has been simulated within the software of MATLAB and it has also been run according to objective functions of SPHERE, GRIEWANK and ACKLEY. All these functions are standard evaluation functions that are generally used for meta-heuristic algorithms. Results that were yielded by the proposed method were better than the results of the initial algorithm and especially by increasing the number of variables of the problem, this improvement becomes even more significant. We have successfully established a better balance between concepts of exploration and exploitation, especially with increasing the repetition cycles, we have successfully controlled the concept of utilization with random parameters. Tests have been ran more than 500 times.