color style


Short Analysis of Implementation and Resource Utilization for the OpenStack Cloud Computing Platform

D. Grzonka, M. Szczygiel, A. Bernasiewicz, A. Wilczynski, M. Liszka: paper and presentation. Proceedings of 29th European Conference on Modelling and Simulation, Albena (Varna), Bulgaria, May 26th – 29th, 2015.

The problem of efficient use of computer resources is the actual challenge for many years. Huge progress in the hardware development has left behind the development of software techniques. One of the most popular solutions for this problem is the idea of virtualization, which a natural continuation is cloud computing. Cloud computing is an innovative concept, where the resources are virtualized, dynamically extended and provided as highly personalized services. In this paper, we present a short analysis of open-source cloud technology - OpenStack. We described OpenStack architecture, requirements, setup process, and related problems. We also conducted a thorough analysis of resource utility - both at full load and without. In our experiments we have analyzed the performance depending on the allocation of virtual resources. Through our work we pointed out aspects which deserve attention by choosing an OpenStack platform. Additionally, we draw attention to the burden on the use of technologies such as OpenStack cloud.

Cloud Computing OpenStack Virtualization High Performance Computing Parallel Environments Resource Utilization Analysis

* * * * *

Artificial Neural Network Support to Monitoring of the Evolutionary Driven Security Aware Scheduling in Computational Distributed Environments

D. Grzonka, J. Kolodziej, J. Tao, S. U. Khan. Future Generation Computer Systems.

Monitoring of the system performance in highly distributed computing environments is a wide research area. In cloud and grid computing, it is usually restricted to the utilization and reliability of the resources. However, in today’s Computational Grids (CGs) and Clouds (CCs), the end users may define the special personal requirements and preferences in the resource and service selection, service functionality and data access. Such requirements may refer to the special individual security conditions for the protection of the data and application codes. Therefore, solving the scheduling problems in modern distributed environments remains still challenging for most of the well known schedulers, and the general functionality of the monitoring systems must be improved to make them efficient as schedulers supporting modules.
In this paper, we define a novel model of security-driven grid schedulers supported by an Artificial Neural Network (ANN). ANN module monitors the schedule executions and learns about secure task-machine mappings from the observed machine failures. Then, the metaheuristic grid schedulers (in our case—genetic-based schedulers) are supported by the ANN module through the integration of the sub-optimal schedules generated by the neural network, with the genetic populations of the schedules.
The influence of the ANN support on the general schedulers’ performance is examined in the experiments conducted for four types of the grid networks (small, medium, large and very large grids), two security scheduling modes—risky and secure scenarios, and six genetic-based grid schedulers. The generated empirical results show the high effectiveness of such monitoring support in reducing the values of the major scheduling criteria (makespan and flowtime), the run times of the schedulers and the grid resource failures.

Distributed environments Artificial Neural Network Monitoring Genetic-based schedulers Scheduling Security awareness Computational grids

* * * * *

Using Artificial Neural Network for Monitoring and Supporting the Grid Scheduler Performance

D. Grzonka, J. Kolodziej, J. Tao: paper and presentation. Proceedings of 28th European Conference on Modelling and Simulation, Brescia, Italy, May 27th – 30th, 2014.

Task scheduling and resource allocations are the key issues for computational grids. Distributed resources usually work at different autonomous domains with their own access and security policies that impact successful job executions across the domain boundaries. In this paper, we propose an Artificial Neural Network (ANN) approach for supporting the security awareness of evolutionary driven grid schedulers. Making a prior analysis of the trust levels of resources and security demand parameters of tasks, the neural network monitors the scheduling and task execution processes. In the result produce the tasks-machines mapping “suggestions”, which can be then utilized by the scheduler to reduce the makespan or increase the system throughput. In this paper, we report the development of risk-resilient genetic-based schedulers and their integration with an ANN module of the HyperSim-G Grid Simulator to evaluate the proposed model under the heterogeneity and large-scale system dynamics. The simulation results showed a significant impact of the ANN support on enhancing the effectiveness of the genetic-based meta-heuristics in reducing the cost of security awareness in grid scheduling.

Computational Grid Scheduling Artificial Neural Network Data Classification Security Genetic Algorithm

* * * * *

Application of High-Performance Techniques for Solving Linear Systems of Algebraic Equations

D. Grzonka. Journal of Telecommunications and Information Technology, 4/2013.

Solving many problems in mechanics, engineering, medicine and other (e.g., diffusion tensor magnetic resonance imaging or finite element modeling) requires the efficient solving of algebraic equations. In many cases, such systems are very complex with a large number of linear equations, which are symmetric positive-defined (SPD). This paper is focused on improving the computational efficiency of the solvers dedicated for the linear systems based on incomplete and noisy SPD matrices by using preconditioning technique – Incomplete Cholesky Factorization, and modern set of processor instructions – Advanced Vector Extension. Application of these techniques allows to fairly reduce the computational time, number of iterations of conventional algorithms and improve the speed of calculation.

Advanced Vector Extension Conjugate Gradient Method Incomplete Cholesky Factorization Preconditioning Vector Registers

* * * * *