Armando Eduardo De Giusti

Particle Swarm Optimization with Variable Population Size

At present, the optimization problem resolution is a topic of great interest, which has fostered the development of several computer methods forsolving them. Particle Swarm Optimization (PSO) is a metaheuristics which has successfully been used in the resolution of a wider range of optimization problems, including neural network training and function minimization. In its original definition, PSO makes use, during the overall adaptive process, of a population made up by a fixed number of solutions. This paper presents a new extension of PSO, called VarPSO, incorporating the concepts of age and neighborhood to allow varying the size of the population. In this way, the quality of the solution to be obtained will not be affected by the used swarms size. The method here proposed is applied to the resolution of some complex functions, finding better results than those typically achieved using a fixed size population.

An energy-aware performance analysis of SWIMM: Smith-Waterman implementation on Intel's Multicore and Manycore architectures

Alignment is essential in many areas such as biological, chemical and criminal forensics. The well‐known Smith–Waterman (SW) algorithm is able to retrieve the optimal local alignment with quadratic time and space complexity. There are several implementations that take advantage of computing parallelization, such as manycores, FPGAs or GPUs, in order to reduce the alignment effort. In this research, we adapt, develop and tune the SW algorithm named SWIMM on a heterogeneous platform based on Intels Xeon and Xeon Phi coprocessor. SWIMM is a free tool available in a public git repository https://github.com/enzorucci/SWIMM. We efficiently exploit data and thread‐level parallelism, reaching up to 380 GCUPS on heterogeneous architecture, 350 GCUPS for the isolated Xeon and 50 GCUPS on Xeon Phi. Despite the heterogeneous implementation obtaining the best performance, it is also the most energy‐demanding. In fact, we also present a trade‐off analysis between performance and power consumption. The greenest configuration is based on an isolated multicore system that exploits AVX2 instruction set architecture reaching 1.5 GCUPS/Watts. Copyright

Smith-Waterman Algorithm on Heterogeneous Systems: A Case Study

The well-known Smith-Waterman (SW) algorithm is a high-sensitivity method for local alignments. However, SW is expensive in terms of both execution time and memory usage, which makes it impractical in many applications. Some heuristics are possible but at the expense of losing sensitivity. Fortunately, previous research have shown that new computing platforms such as GPUs and FPGAs are able to accelerate SW and achieve impressive speedups. In this paper we have explored SW acceleration on a heterogeneous platform equipped with an Intel Xeon Phi coprocessor. Our evaluation, using the well-known Swiss-Prot database as a benchmark, has shown that a hybrid CPU-Phi heterogeneous system is able to achieve competitive performance (62.6 GCUPS), even with moderate low-level optimisations.

A new binary PSO with velocity control

Particle Swarm Optimization (PSO) is a metaheuristic that is highly used to solve mono- and multi-objective optimization problems. Two well-differentiated PSO versions have been defined - one that operates in a continuous solution space and one for binary spaces. In this paper, a new version of the Binary PSO algorithm is presented. This version improves its operation by a suitable positioning of the velocity vector. To achieve this, a new modified version of the continuous gBest PSO algorithm is used. The method proposed has been compared with two alternative methods to solve four known test functions. The results obtained have been satisfactory.

Dynamic on Demand Virtual Clusters in Grid

In Grid environments, many different resources are intended to work in a coordinated manner, each resource having its own features and complexity. As the number of resources grows, simplifying automation and management is among the most important issues to address. This papers contribution lies on the extension and implementation of a grid metascheduler that dynamically discovers, creates and manages on-demand virtual clusters. The first module selects the clusters using graph heuristics. The algorithm then tries to find a solution by searching a set of clusters, mapped to the graph, that achieve the best performance for a given task. The second module, one per-grid node, monitors and manages physical and virtual machines. When a new task arrives, these modules modify virtual machines configuration or use live migration to dynamically adapt resource distribution at the clusters, obtaining maximum utilization. Metascheduler components and local administrator modules work together to make decisions at run time to balance and optimize system throughput. This implementation results in performance improvement of 20% on the total computing time, with machines and clusters processing 100% of their working time. These results allow us to conclude that this solution is feasible to be implemented on Grid environments, where automation and self-management are key to attain effective resource usage.

Learning Life Cycle in Autonomous Intelligent Systems

Autonomous Intelligent Systems (AIS) integrate planning, learning, and execution in a closed loop, showing an autonomous intelligent behavior. A Learning Life Cycle (LLC) for AISs is proposed. The LLC is based on three different learned operators’ layers: Built-in Operators, Trained Base Operators and World Interaction Operators. The extension of the original architecture to support the new type of operators is presented.

Dynamic load balancing on non-homogeneous clusters

This paper discusses the dynamic and static balancing of non-homogenous cluster architectures, simultaneously analyzing the theoretical parallel speedup as well as the speedup experimentally obtained. A classical application (Parallel N-Queens) with a parallel solution algorithm, where processing predominates upon communication, has been chosen so as to go deep in the load balancing aspects (dynamic or static) without distortion of results caused by communication overhead. Four interconnected clusters have been used in which the machines within each cluster have homogeneous processors although different among clusters. Thus, the set can be seen as a N-processor heterogeneous cluster or as a multi-cluster scheme with 4 subsets of homogeneous processors. At the same time, three forms of load distribution in the processors (Direct Static, Predictive Static and Dynamic by Demand) have been studied, analyzing in each case parallel speedup and load unbalancing regarding problem size and the processors used.

OSWALD: OpenCL Smith–Waterman on Altera’s FPGA for Large Protein Databases

The well-known Smith–Waterman algorithm is a high-sensitivity method for local sequence alignment. Unfortunately, the Smith–Waterman algorithm has quadratic time complexity, which makes it computationally demanding for large protein databases. In this paper, we present OSWALD, a portable, fully functional and general implementation to accelerate Smith–Waterman database searches in heterogeneous platforms based on Altera’s FPGA. OSWALD exploits OpenMP multithreading and SIMD computing through SSE and AVX2 extensions on the host while taking advantage of pipeline and vectorial parallelism by way of OpenCL on the FPGAs. Performance evaluations on two different heterogeneous architectures with real amino acid datasets show that OSWALD is competitive in comparison with other top-performing Smith–Waterman implementations, attaining up to 442 GCUPS peak with the best GCUPS/watts ratio.The well-known Smith–Waterman algorithm is a high-sensitivity method for local sequence alignment. Unfortunately, the Smith–Waterman algorithm has quadratic time complexity, which makes it computat...

Rule Extraction on Numeric Datasets Using Hyper-rectangles

When there is a need to understand the data stored in a database, one of the main requirements is being able to extract knowledge in the form of rules. Classification strategies allow extracting rules almost naturally. In this paper, a new classification strategy is presented that uses hyper-rectangles as data descriptors to achieve a model that allows extracting knowledge in the form of classification rules. The participation of an expert for training the model is discussed. Finally, the results obtained using the databases from the UCI repository are presented and compared with other existing classification models, showing that the algorithm presented requires less computational resources and achieves the same accuracy level and number of extracted rules.

VarMOPSO: multi-objective particle swarm optimization with variable population size

The PSO (Particle Swarm Optimization) metaheuristics, originally defined for solving single-objective problems, has been applied to multi-objective problems with very good results. In its initial conception, the algorithm has a fixed-size population. In this paper, a new variation of this metaheuristics, called VarMOPSO (Variable Multi-Objective Particle Swarm Optimization), characterized by a variable-sized population, is proposed. To this end, the concepts of age and neighborhood are incorporated to be able to modify the size of the population for the different generations. This algorithm was compared with the version that uses fixed-size populations, as well as with other metaheuristics, all of them representative of the state of the art in multi-objective optimization. In all cases, three widely used metrics were considered as quality indicators for Pareto front. The results obtained were satisfactory.