Comparative study of algorithms for mining association rules: Traditional approach versus multi-relational approach

The multi-relational Data Mining approach has emerged as alternative to the analysis of structured data, such as relational databases. Unlike traditional algorithms, the multi-relational proposals allow mining directly multiple tables, avoiding the costly join operations. In this paper, is presented a comparative study involving the traditional Patricia Mine algorithm and its corresponding multi-relational proposed, MR-Radix in order to evaluate the performance of two approaches for mining association rules are used for relational databases. This study presents two original contributions: the proposition of an algorithm multi-relational MR-Radix, which is efficient for use in relational databases, both in terms of execution time and in relation to memory usage and the presentation of the empirical approach multirelational advantage in performance over several tables, which avoids the costly join operations from multiple tables. © 2011 IEEE.

Multi-relational algorithm for mining association rules in large databases

Multi-relational data mining enables pattern mining from multiple tables. The existing multi-relational mining association rules algorithms are not able to process large volumes of data, because the amount of memory required exceeds the amount available. The proposed algorithm MRRadix presents a framework that promotes the optimization of memory usage. It also uses the concept of partitioning to handle large volumes of data. The original contribution of this proposal is enable a superior performance when compared to other related algorithms and moreover successfully concludes the task of mining association rules in large databases, bypass the problem of available memory. One of the tests showed that the MR-Radix presents fourteen times less memory usage than the GFP-growth. © 2011 IEEE.

Optimization of algorithm to identification of duplicate tuples through similarity phonetic based on multithreading

Aiming to ensure greater reliability and consistency of data stored in the database, the data cleaning stage is set early in the process of Knowledge Discovery in Databases (KDD) and is responsible for eliminating problems and adjust the data for the later stages, especially for the stage of data mining. Such problems occur in the instance level and schema, namely, missing values, null values, duplicate tuples, values outside the domain, among others. Several algorithms were developed to perform the cleaning step in databases, some of them were developed specifically to work with the phonetics of words, since a word can be written in different ways. Within this perspective, this work presents as original contribution an optimization of algorithm for the detection of duplicate tuples in databases through phonetic based on multithreading without the need for trained data, as well as an independent environment of language to be supported for this. © 2011 IEEE.

Optimization of query results in peer-to-peer database supported by semantics

The development of new technologies that use peer-to-peer networks grows every day, with the object to supply the need of sharing information, resources and services of databases around the world. Among them are the peer-to-peer databases that take advantage of peer-to-peer networks to manage distributed knowledge bases, allowing the sharing of information semantically related but syntactically heterogeneous. However, it is a challenge to ensure the efficient search for information without compromising the autonomy of each node and network flexibility, given the structural characteristics of these networks. On the other hand, some studies propose the use of ontology semantics by assigning standardized categorization of information. The main original contribution of this work is the approach of this problem with a proposal for optimization of queries supported by the Ant Colony algorithm and classification though ontologies. The results show that this strategy enables the semantic support to the searches in peer-to-peer databases, aiming to expand the results without compromising network performance. © 2011 IEEE.

A logical approach to the management of object identifiers in non-conventional database management systems

Non-conventional database management systems are used to achieve a better performance when dealing with complex data. One fundamental concept of these systems is object identity (OID), because each object in the database has a unique identifier that is used to access and reference it in relationships to other objects. Two approaches can be used for the implementation of OIDs: physical or logical OIDs. In order to manage complex data, was proposed the Multimedia Data Manager Kernel (NuGeM) that uses a logical technique, named Indirect Mapping. This paper proposes an improvement to the technique used by NuGeM, whose original contribution is management of OIDs with a fewer number of disc accesses and less processing, thus reducing management time from the pages and eliminating the problem with exhaustion of OIDs. Also, the technique presented here can be applied to others OODBMSs. © 2011 IEEE.

Spatial clustering applied to health area

The significant volume of work accidents in the cities causes an expressive loss to society. The development of Spatial Data Mining technologies presents a new perspective for the extraction of knowledge from the correlation between conventional and spatial attributes. One of the most important techniques of the Spatial Data Mining is the Spatial Clustering, which clusters similar spatial objects to find a distribution of patterns, taking into account the geographical position of the objects. Applying this technique to the health area, will provide information that can contribute towards the planning of more adequate strategies for the prevention of work accidents. The original contribution of this work is to present an application of tools developed for Spatial Clustering which supply a set of graphic resources that have helped to discover knowledge and support for management in the work accidents area. © 2011 IEEE.

Managing object identity with less I/O operations

Non-conventional database management systems are used to achieve a better performance when dealing with complex data. One fundamental concept of these systems is object identity (OID). Two techniques can be used for the implementation of OIDs: physical or logical. A logical implementation of OIDs, based on an Indirection Table, is used by NuGeM, a multimedia data manager kernel which is described in this paper. NuGeM Indirection Table allows the relocation of all pages in a database. The proposed strategy modifies the workings of this table so that it is possible to reduce considerably the number of I/O operations during the request and release of pages containing objects and their OIDs. Tests show a reduction of 84% in reading operations and a 67% reduction in writing operations when pages are requested. Although no changes were observed in writing operations during the release of pages, a 100% of reduction in reading operations was obtained. © 2012 IEEE.

GPGPU implementation of growing neural gas: application to 3D scene reconstruction

Self-organising neural models have the ability to provide a good representation of the input space. In particular the Growing Neural Gas (GNG) is a suitable model because of its flexibility, rapid adaptation and excellent quality of representation. However, this type of learning is time-consuming, especially for high-dimensional input data. Since real applications often work under time constraints, it is necessary to adapt the learning process in order to complete it in a predefined time. This paper proposes a Graphics Processing Unit (GPU) parallel implementation of the GNG with Compute Unified Device Architecture (CUDA). In contrast to existing algorithms, the proposed GPU implementation allows the acceleration of the learning process keeping a good quality of representation. Comparative experiments using iterative, parallel and hybrid implementations are carried out to demonstrate the effectiveness of CUDA implementation. The results show that GNG learning with the proposed implementation achieves a speed-up of 6× compared with the single-threaded CPU implementation. GPU implementation has also been applied to a real application with time constraints: acceleration of 3D scene reconstruction for egomotion, in order to validate the proposal.

Large scale data processing using MapReduce

As massive data sets become increasingly available, people are facing the problem of how to effectively process and understand these data. Traditional sequential computing models are giving way to parallel and distributed computing models, such as MapReduce, both due to the large size of the data sets and their high dimensionality. This dissertation, as in the same direction of other researches that are based on MapReduce, tries to develop effective techniques and applications using MapReduce that can help people solve large-scale problems. Three different problems are tackled in the dissertation. The first one deals with processing terabytes of raster data in a spatial data management system. Aerial imagery files are broken into tiles to enable data parallel computation. The second and third problems deal with dimension reduction techniques that can be used to handle data sets of high dimensionality. Three variants of the nonnegative matrix factorization technique are scaled up to factorize matrices of dimensions in the order of millions in MapReduce based on different matrix multiplication implementations. Two algorithms, which compute CANDECOMP/PARAFAC and Tucker tensor decompositions respectively, are parallelized in MapReduce based on carefully partitioning the data and arranging the computation to maximize data locality and parallelism.

Multi-Tenant Virtual GPUs for Optimising Performance of a Financial Risk Application

Graphics Processing Units (GPUs) are becoming popular accelerators in modern High-Performance Computing (HPC) clusters. Installing GPUs on each node of the cluster is not efficient resulting in high costs and power consumption as well as underutilisation of the accelerator. The research reported in this paper is motivated towards the use of few physical GPUs by providing cluster nodes access to remote GPUs on-demand for a financial risk application. We hypothesise that sharing GPUs between several nodes, referred to as multi-tenancy, reduces the execution time and energy consumed by an application. Two data transfer modes between the CPU and the GPUs, namely concurrent and sequential, are explored. The key result from the experiments is that multi-tenancy with few physical GPUs using sequential data transfers lowers the execution time and the energy consumed, thereby improving the overall performance of the application.

Predictive and distributed routing balancing (PR-DRB): high speed interconnection networks

Current parallel applications running on clusters require the use of an interconnection network to perform communications among all computing nodes available. Imbalance of communications can produce network congestion, reducing throughput and increasing latency, degrading the overall system performance. On the other hand, parallel applications running on these networks posses representative stages which allow their characterization, as well as repetitive behavior that can be identified on the basis of this characterization. This work presents the Predictive and Distributed Routing Balancing (PR-DRB), a new method developed to gradually control network congestion, based on paths expansion, traffic distribution and effective traffic load, in order to maintain low latency values. PR-DRB monitors messages latencies on intermediate routers, makes decisions about alternative paths and record communication pattern information encountered during congestion situation. Based on the concept of applications repetitiveness, best solution recorded are reapplied when saved communication pattern re-appears. Traffic congestion experiments were conducted in order to evaluate the performance of the method, and improvements were observed.

Extension and evaluation of JXTA protocols for supporting reliable P2P distributed computing

Peer-reviewed

An online data access prediction and optimization approach for distributed systems

Current scientific applications have been producing large amounts of data. The processing, handling and analysis of such data require large-scale computing infrastructures such as clusters and grids. In this area, studies aim at improving the performance of data-intensive applications by optimizing data accesses. In order to achieve this goal, distributed storage systems have been considering techniques of data replication, migration, distribution, and access parallelism. However, the main drawback of those studies is that they do not take into account application behavior to perform data access optimization. This limitation motivated this paper which applies strategies to support the online prediction of application behavior in order to optimize data access operations on distributed systems, without requiring any information on past executions. In order to accomplish such a goal, this approach organizes application behaviors as time series and, then, analyzes and classifies those series according to their properties. By knowing properties, the approach selects modeling techniques to represent series and perform predictions, which are, later on, used to optimize data access operations. This new approach was implemented and evaluated using the OptorSim simulator, sponsored by the LHC-CERN project and widely employed by the scientific community. Experiments confirm this new approach reduces application execution time in about 50 percent, specially when handling large amounts of data.

Design and Development of an FPGA-based Distributed Computing Processing Platform

This thesis presents two frameworks- a software framework and a hardware core manager framework- which, together, can be used to develop a processing platform using a distributed system of field-programmable gate array (FPGA) boards. The software framework providesusers with the ability to easily develop applications that exploit the processing power of FPGAs while the hardware core manager framework gives users the ability to configure and interact with multiple FPGA boards and/or hardware cores. This thesis describes the design and development of these frameworks and analyzes the performance of a system that was constructed using the frameworks. The performance analysis included measuring the effect of incorporating additional hardware components into the system and comparing the system to a software-only implementation. This work draws conclusions based on the provided results of the performance analysis and offers suggestions for future work.