Pradeep Kumar Yadav

Dynamic Tasks Scheduling Model for Performance Evaluation of a Distributed Computing System through Artificial Neural Network

As technology has quickly and relentlessly advanced in the field of computer hardware, Distributed Computing System [DCS] has become increasingly popular. Performance enhancement is one of the most important issues in distributed systems. In this paper we have proposed a dynamic task allocation model based on artificial neural network [ANN] scheduling approach to arrange the tasks to the processors. Relocating the tasks from one processor to another at certain points during the course of execution of the program that contributes to the total cost of the running program has been taken into account. Phase-wise Execution Cost [EC], Inter Task Communication Cost [ITCC], Residence Cost [RC] of each task on different processors and Relocation Cost [REC] for each task has been considered while preparing the model.

Dynamic Tasks Scheduling Algorithm for Distributed Computing Systems under Fuzzy Environment

Distributed computing systems [DCS] offer the potential for allocating a number of tasks to different processors for execution. It is desired to assign the tasks dynamically to that processor whose characteristics are most appropriate for the execution in order to make the best use of the computational power available. This paper proposes a new mathematical model for allocating the tasks of distributed program to multiple processors in order to achieve optimal cost and optimal reliability of the system. Phase-wise execution cost, residence cost of each task on different processors, inter task communication cost and relocation cost for each task have been considered as a fuzzy number which is more realistic and general in nature. The fuzzy problem has been transformed into crisp one by using the defuzzification method. The present algorithm is formulated and applied to numerical examples to demonstrate its effectiveness. The present model is suitable for arbitrary number of phases and processors with random program structure.

Reliability Driven Soft Real-Time Fuzzy Task Scheduling in Distributed Computing Environment

All practical real-time scheduling algorithms in distributed systems present a trade-off between their computational complexity and performance. In real-time distributed systems, tasks have to be performed correctly, timely and most importantly in reliable manner. The research till date on task scheduling has primarily focused upon computation time, laxity, priority etc. Reliability of a task or task cluster is normally not considered, when it comes to task scheduling in distributed environment. Finding a reliable schedule in distributed systems with real-time constraints is shown to be NP-hard. The practical scheduling algorithms in real-time systems have not deterministic reliability. Deterministic reliable behavior is an important parameter for system robustness analysis. The intrinsic uncertainty in dynamic real-time systems increases the difficulties of scheduling problem. To alleviate these deficiencies, i have proposed a fuzzy scheduling approach to arrange real-time periodic and non-periodic tasks in systems considering reliability as the parameter for decision. In contrast, the approach balances task loads of the processors successfully while consider starvation prevention and fairness, which cause higher reliability tasks, have higher running probability. A simulation is conducted to evaluate the performance of the proposed approach. Experimental results have shown that the proposed fuzzy scheduler creates feasible schedules for homogeneous and heterogeneous tasks.

GA-Based Task Scheduling Algorithm for Efficient Utilization of Available Resources in Computational Grid

In the grid computing environment, systematic scheduling of tasks/jobs on hand resource is the important parameter for performance evaluation of computational grid. Traditional algorithms cannot produce a load balancing schedule. In the paper, a genetic approach for grid task scheduling has been considered to achieve better solutions within a reasonable period of time. The present study aims at minimizing the make-span and flow-time at the same time and also achieves equiponderant practical application of a set of “n” available computing agents of a grid computing to get the average load balancing. The simulation results show that the proposed approach is more efficient than the GA approach reported in the literature.

A task allocation model for minimising system cost and maximising reliability of distributed computing system

A distributing computing system (DCS) presents a platform consisting of multiple computing nodes connected in some fashion to which various task of a parallel application could be assigned. Effective task allocation is an essential phase to obtain the optimal cost and reliability of DCS. This paper addressed a new heuristic model of allocating the tasks onto multiple processors in order to achieve optimal cost and optimal reliability of the system. Costs have been taken as fuzzy numbers, which are more realistic and general in nature. Fuzzy program is transformed into crisp one by using the method of defuzzification. A membership function is developed to transform the communication cost in unit interval [0, 1]. Clusters of tasks have been built on the basis of their membership values. To present the effect of the proposed model, it has been implemented on various numerical examples and compared with other existing method related to this field.

A Fuzzy Clustering Method for Performance Evaluation for Distributed Real Time Systems

All practical real-time scheduling algorithm in distributed processing environment present a trade-off between their computational intricacy and performance. In real-time system, tasks have to perform correctly and timely. Finding minimal schedule in distributed processing system with constrains is shown to be NP- Hard. Systematic allocation of task in distributed real-time system is one of the major important parameter to evaluate the performance if this step is not execute properly the throughput of the system may be decrease. In this paper, a fuzzy clustering-based algorithm has been discussed.

Workload Analysis in a Grid Computing Environment: A Genetic Approach

Grid computing is the collection of computer resources from multiple locations to reach a common goal. The grid is a special type of distributed system with non-interactive workloads that involve a large number of files. Partitioning of the application program/ software into a number of small groups of modules among dissimilar processors is an important parameter to determine the efficient utilization of available resources in a grid computing environment. It also enhances the computation speed. The task partitioning and task allocation activities influence the distributed program/ software properties such as IPC. This paper presents a metaheuristic model, that performs static allocation of a set of “m” modules of distributed tasks/program considering the two conflicting objectives i.e. minimizing the makespan time and balanced utilization of a set of “n” available resources of a grid computing. Experimental results using genetic algorithm indicates that the proposed algorithm achieved these two objectives as well as improve the dynamic heuristics presented in literature.

A Secure SCAM (Smart Card based Authentication Mechanism)

Recently, Tsai et al. proposed dynamic ID based smart card authentication scheme. This paper demonstrates that Tsai et al.’s scheme fails to provide early wrong password detection, secure password change and protection against insider attack. To overcome, we propose a secure SCAM (Smart Card based Authentication Mechanism) which keeps all previous merits and achieves security and functionality requirements. The performance of both the schemes has been analyzed in terms of various metrics. Comparing with Tsai et al.’s scheme, our scheme provides higher security with nearly same cost. For network where clock synchronization is tough, nonce based scheme is additionally offered.

Validation and Optimization of Stress Induced using Ansys

-It is difficult to determine the mechanical properties of composite materials like stress,strain etc due to time consumption and extra expenditure in proper mechanical testing and designing.So once methodology is established by using software Ansys , that process will be less time consuming and of less expenditure. Methodology will be established by validating software analysis with the industrial data and then optimized using various resin.