Abstract Service identification (SI) in the life cycle of service‐oriented architecture is a critical phase. Business models consisting of business process (BP) model and business entity (BE) model are the useful models that may be used for SI. To this end, SI is carried out by partitioning activities in BP based on the activities' use of the entities in BE. However, a proper partitioning activities to services, which is called a service design, is a challenge. This article aims to present a semiautomatized clustering method for partitioning the activities to services, which is directed by new proposed metrics cohesion, coupling, and granularity. With regard to the conflict of the metrics, a multiobjective evolutionary algorithm (MOEA) is used to clustering activities where the metrics are considered as objectives should be optimized. The MOEA produces a set of optimal solutions as proper identified services of a service design. Finally, we used three case studies to show the effectiveness of the proposed method and then evaluated the results.

Summary Allocation of shared resources by multithreaded programs faces problem of deadlock. Many solutions have been presented to resolve this problem. Among others, the deadlock prevention is stressed when the deadlock detection and removal are costly. Removing deadlocks, which is carried out by aborting/recovering deadlocked threads, causes the waste of the resources used by the deadlocked threads. In such case, deadlock prevention can lead to avoiding the waste of resources. To this end, the runtime behavior of threads should be monitored in order to predict possible future deadlocks. In such case, the prediction mechanism becomes significant because a proper prediction helps us deny the allocation request of a resource by a thread if the allocation leads to a potential deadlock. A method to attain to proper prediction is learning the behavior of threads based on runtime monitoring their past behavior. In fact, based on past behavior of threads, potential deadlocks in their future behavior are verified, and current allocation request of a resource is denied if a future deadlock is predicted. The current study is an extension of our previous work where just deadlock tracking was predicted and no adaptation was suggested. In this study, a composite structure of a recurrent Neural Network (NN) called NARX (to track a potential deadlock) and a Multi‐perceptron NN called MLP (to select a suitable action to resolve the potential deadlock) is proposed. Based on the experimental results, the accuracy of the first NN was about 80%, leading to high performance of the second NN, and more than 82% of the real deadlocks were prevented by selecting suitable actions.

A pacemaker is an embedded device that is sited in the chest to regulate irregular heartbeats known as arrhythmias. Since such devices are directed by software, a software failure may cause a serious hazard such as patient death. Runtime monitoring and adaptation of the device software behavior offers a solution for preventing device hazards. We have already obtained some experiences from monitoring medical devices like: 1) insulin pump using fuzzy Petri net and 2) cardiac pacemaker using colored Petri-net (CPN) and hierarchical fuzzy CPN (HFCPN). However, these studies did not present an adaptation method for software runtime faults. This paper, extending our previous work, presents an automatic runtime method for continuous verification of the behavior of the implanted pacemaker using a software agent. The autonomy and the intelligence characteristics of the software agent are used to control the behavior of the cardiac pacemaker software by drawing inferences from a knowledge base where HFCPN is used by the agent as the inference engine. Compared to a flat inference engine, the HFCPN is able to cover the concurrent states initiated by input fuzzy values and improve the running time for finding a suitable rule by up to 92%. In addition, the intelligent software agent checks the runtime operation accuracy of the pacemaker software in vital and unexpected situations, and redirects the software decision if it finds an unacceptable value. To demonstrate the HFCPN's behavior and decision-making in different situations, three different scenarios are presented.

Summary Dynamic on‐demand resource provisioning is one of the primary goals of the cloud computing task scheduling process. Task scheduling is a nondeterministic polynomial time (NP)‐hard problem and is responsible for assigning tasks to virtual machines (VMs) in a way that increases the resource utilization and performance, reduces response time, and keeps the whole system balanced. In this paper, we present a static task scheduling method based on the particle swarm optimization (PSO) algorithm where the tasks are assumed to be non‐preemptive and independent. We have improved the performance of the basic PSO method using a load‐balancing technique. We have compared our proposed method with round robin (RR) task scheduling, improved PSO task scheduling and a load‐balancing technique. The simulation results show that our method outperforms these algorithms by an increase of resource utilization of 22% and a decrease of makespan by 33%, compared with the basic PSO algorithm. The results illustrate that our proposed method converges to the near optimal solution faster than the basic PSO algorithm and is more efficacious with more tasks.

Summary Because of widespread distribution of resources in the geographically distributed cloud environment, optimal selection of virtual machines (VMs) is one of the most important challenges for the structure of the network. This is due to the high number of data centers and VMs with different qualities of service parameters. Because of redundancy in the VMs and the high number of service parameters, optimal selection of VMs is an NP‐hard problem. Therefore, a method is required, which can suggest the best VMs on the basis of the user's request and on the service‐level agreements (SLAs). This study focuses on four important factors in SLAs: cost, response time, availability, and reliability. In this paper, we propose a four‐tier structure, Observe, Orient, Decide, and Act, where (1) Observe is responsible for continuous monitoring users' requests and characteristics of data centers and VMs, (2) Orient is responsible for clustering data centers using fuzzy c‐means and based of the four quality of services (SLA's factors) and then the selection of the most suitable data center cluster for the VM selection, (3) Decide is responsible for making decision on the most suitable VMs using multiobjective linear programming, and (4) Act is responsible for the execution of the decision. The proposed structure was implemented, and its effectiveness was evaluated through considering the number of SLA violations.

Summary A workflow consists of dependent tasks, and scheduling of a workflow in a cloud environment means the arrangement of tasks of the workflow on virtual machines (VMs) of the cloud. By increasing VMs and the diversity of task size, we have a huge number of such arrangements. Finding an arrangement with minimum completion time among all of the arrangements is an Non‐Polynomial‐hard problem. Moreover, the problem becomes more complex when a scheduling should consider a couple of conflicting objectives. Therefore, the heuristic algorithms have been paid attention to figure out an optimal scheduling. This means that although the single‐objective optimization, ie, minimizing completion time, proposes the workflow scheduling as an NP‐complete problem, multiobjective optimization for the scheduling problem is confronted with a more permutation space because an optimal trade‐off between the conflicting objectives is needed. To this end, we extended a recent heuristic algorithm called Grey Wolf Optimizer (GWO) and considered dependency graph of workflow tasks. Our experiment was carried out using the WorkflowSim simulator, and the results were compared with those of 2 other heuristic task scheduling algorithms.

Abstract Failure of deterministic encryption algorithms against various attacks mostly occurs due to the use of fixed and deterministic methods for encrypting, decrypting, and key production. Because the key production in these algorithms is performed functionally and deterministically, there is a relation between the keys and their length that makes the differential analysis of the keys possible. Moreover, the fixed length and values of the keys make these algorithms unsafe against exhaustive key searches. In this study, a block cipher algorithm called indefinite block ciphering based on variable and great length key which belongs to probabilistic (nondeterministic) family is presented. In this algorithm, the key depends on the plain text and is produced indecisively by a random number and the connectivity of the different parameters. On the other hand, the plain text is divided into blocks with size of random number. The proposed algorithm was evaluated, and the results were compared with those of the deterministic and other nondeterministic algorithms. This comparison was indicative of the high performance and security of the new algorithm with respect to the other approaches. Copyright © 2016 John Wiley & Sons, Ltd.

Summary In a multithreaded program, competition of threads for shared resources raises the deadlock possibility, which narrows the system liveness. Because such errors appear in specific schedules of concurrent executions of threads, runtime verification of threads behavior is a significant concern. In this study, we extended our previous approach for prediction of runtime behavior of threads may lead to an impasse. Such a prediction is of importance because of the nondeterministic manner of competing threads. The prediction process tries to forecast future behavior of threads based on their observed behavior. To this end, we map observed behavior of threads into time‐series data sets and use statistical and artificial intelligence methods for forecasting subsequent members of the sets as future behavior of the threads. The deadlock prediction is carried out based on probing the allocation graph obtained from actual and predicted allocation of resources to threads. In our approach, we use an artificial neural network (ANN) because ANNs enjoy the applicable performance and flexibility in predicting complex behavior. Using three case studies, we contrasted results of the current and our previous approaches to demonstrate results. Copyright © 2015 John Wiley & Sons, Ltd.

This paper aims to present a systematic method to: (1) specify high-level and event based real-time requirements and (2) map the specified requirements to low-level and state-based one. The former indicates the external system behavior while the latter indicates the internal one, which the external behavior are specified in environment events and the internal behavior is specified in software entities and operations such as variables and method calls. The mapping can be used in software development process and software monitoring against safety requirements. Lastly, we apply our method to requirements of a real-time safety critical system called Railroad Crossing Control (RCC).