Abstract: The importance of distributed applications is constantly rising due to technological trends such as the widespread usage of smart phones and the increasing internetworking of all kinds of devices. In addition to classical application scenarios with a rather static structure these trends push forward dynamic settings, in which service providers may continuously vanish and newly appear. In this paper categories of distributed applications are identified and analyzed with respect to their most important development challenges. In order to tackle these problems already on a conceptual level the active component paradigm is proposed, bringing together ideas from agents, services and components using a common conceptual perspective. Besides conceptual foundations of active components also a programming model and an implemented infrastructure are presented. It is highlighted how active components help addressing the initially posed challenges by presenting several real world example applications.

Abstract: The 4CaaSt project aims at developing a PaaS framework that enables flexible definition, marketing, deployment and management of Cloud-based services and applications. This paper describes the major challenges tackled by 4CaaSt for the comprehensive management of applications and services in a PaaS. These challenges involve the blueprint language to describe applications in the cloud and its lifecycle management, as well as a one stop shop for Cloud services and a PaaS level resource management featuring elasticity and advanced Network as a Service capabilities. 4CaaSt also provides a portfolio of ready to use Cloud native services and Cloud enabled immigrant technologies. The evaluation process followed to assess 4CaaSt progress is also described.

Abstract: This paper presents an approach to evaluate the supply chain agility behaviour consisting in the development of an integrated index, with the data gathering, transmission and processing supported by a cloud-computing environment The proposed approach relies on the development of two agility indices: one to assess the individual company agile behaviour, and the other one to determine the same behaviour for the entire supply chain The supply chain is presented as a living, self-organizing open system that has the ability to incorporate new efficient agents and to remove the weakest ones A special emphasis is given to the living subsystem responsible for the agility assessment, namely regarding the conceptual details of the components necessary to gather, process, coordinate and control the flow of information in the cloud

Abstract: Recently, Cloud Computing has become an emerging research topic in response to the shift from product-oriented economy to service-oriented economy and the move from focusing on software/system development to addressing business-IT alignment. From the IT perspectives, there is a proliferation of methods for cloud application development. Such methods have clearly shown considerable shortcomings to provide an efficient solution to deal with major aspects related to cloud applications. One of these major aspects is the multi-tenancy of the Software-as-a-Service (SaaS) components used to compose Service-Based Applications (SBAs) on the cloud. Current SaaS offerings are often provided as monolithic one-size-fits-all solutions and give little or no opportunity for further customization. Monolithic SaaS offerings are more likely to show failure in meeting the business requirements of several consumers. In this paper, we analyze the state-of-the-art of the standardization, methodology, software and product support for SBA development on the cloud, identify some shortcomings, and point out the need of a novel approach for breaking down the monolithic stack of cloud service offerings and providing an effective and flexible solution for SBA designers to select, customize, and aggregate cloud service offerings coming from different providers (25)

Abstract: This article is focused on application of agile principles during adoption of Business Process Management in an organization.

Abstract: Graphics Processing Units (GPUs) are massive data parallel processors. High performance comes only at the cost of identifying data parallelism in the applications while using data parallel processors like GPU. This is an easy effort for applications that have regular memory access and high computation intensity. GPUs are equally attractive for sparse matrix vector multiplications (SPMV for short) that have irregular memory access. SPMV is an important computation in most of the scientific and engineering applications and scaling the performance, bandwidth utilization and compute intensity (ratio of computation to the data access) of SPMV computation is a priority in both academia and industry. There are various data structures and access patterns proposed for sparse matrix representation on GPUs and optimizations and improvements on these data structures is a continuous effort. This paper proposes a new format for the sparse matrix representation that reduces the data organization time and the memory transfer time from CPU to GPU for the memory bound SPMV computation. The BLSI (Bit Level Single Indexing) sparse matrix representation is up to 204\% faster than COO (Co-ordinate), 104\% faster than CSR (Compressed Sparse Row) and 217\% faster than HYB (Hybrid) formats in memory transfer time from CPU to GPU. The proposed sparse matrix format is implemented in CUDA-C on CUDA (Compute Unified Device Architecture) supported NVIDIA graphics cards.

Abstract: While adopting the Cloud, the small and medium sized enterprises (SMEs) could increase their benefit by embracing some Platform-as-a-Service (PaaS) solutions, doubled by a Cloud Governance or Cloud brokerage approach. However, in order to fully exploit the advantages brought by Cloud environments, and enter in real competition with "big players" from different markets, SMEs must group themselves under the umbrella of a common marketplace, via some Cloud Governance solutions, and expose together complex, tailored and integrated solutions. The implementation of an effective Cloud Governance solution requires a strong support for storing and manipulating data which is relevant for various aspects of applications, both at business and technical level, support which is closely linked to cloud service lifecycle. This paper focuses on analyzing the main requirements related to data storage in an effective Cloud Governance system, details the functionality of the most important datastores and presents various use cases involving the entire storage architecture.

Abstract: Cloud computing paradigm has attracted a lot of attention in the last five years as coming during an economic crisis with an appealing offer in reducing the infrastructure and maintenance costs. After first wave of enthusiasm in adopting the concept, a clearer image has been formed about the benefits and limitations of Cloud computing and a lot of different supporting technologies were developed. As consequence a new threat is raised by the high number of the proprietary technologies that makes difficult the decision of the proper technological selection according to the real business needs. In this context the aim of this paper is to offer a snapshot on the current concepts and the available technologies, especially of the ones that can allow the development of a solid market of Cloud services and applications. A particular attention is given to the trend of federating and brokering Cloud services in the process of forming new markets. Moreover, we propose a classification of groups of services from multiple Clouds based on models similar to the ones used in computer graphics to express colors. Furthermore, a technological solution aligned to the market requirements is presented as case study, pointing also to the role of open-source codes for promoting the Cloud service usage on large scale.

Abstract: Although Multi Agent Systems (MAS) have attracted a great deal of attention in the field of software engineering, with their promise of capturing complex systems, they remain far away from commercial popularity mainly due to the accessibility of MAS methodologies for commercial developers. In this paper we present a practical method for developing self-managing MAS that we believe enables not only software developers but also business people beyond the academic community to design and develop MAS using familiar concepts. We present the main three phases of the proposed methodology, with details and examples of all the visual models, followed by details of its supporting metamodel, in which we describes the MAS concepts used and their relationships. In particular, the methodology features 1. a formal specification mechanism for system norms 2. offers organizational support of MAS through institutions, and 3. supports self-management explicitly through dynamic planning.

Abstract: Members of bacterial communities communicate and cooperate via diffusible chemical materials they emit into the environment, and at the same time, they also compete for nutrients and space. Agent-based models (ABMs) are useful tools for simulating the growth of communities containing multiple interacting microbial species. In this work we present numerical indices characterizing spatial distribution and the fitness of competing bacterial species in an ABM and we present data on how these indices can be used to visually summarize large scale simulation experiments. Preliminary results show bacterial agents utilizing different nutrients but sharing communication signals and public goods can form stable mixed communities in which the species grow faster than any of the single species alone.

Abstract: In this article, a cognitive management framework is proposed for ensuring exploitation of the Future Internet of Things (FIoT). Cognitive systems offer self-x and learning. A cognitive system has the ability to dynamically select its behavior through self-management/awareness functionality, taking into account information and knowledge on the context of the operations as well as policies and including the generation of the context itself. The framework is based on the principle that any real world object and any digital object that is available, accessible, observable or controllable can have a virtual representation in the Future Internet, which is called Virtual Object (VO). Basic VOs can be composed in a more sophisticated way by forming Composite VOs (CVOs), which provide services to high-level applications and end-users. The described paradigm is applied to various applications scenarios: smart home, smart office, smart city and smart business. This paper presents some background in IoT, identifies the requirements and challenges, and sets the directions that should be followed. © 2012 SCPE.

Abstract: The paper concerns designing distributed program execution control based on global application states monitoring in the presence of a dynamic number of processes and threads. Global program execution control is based on application states monitoring at the level of processes/threads in clusters of multi--core processors. A special control infrastructure is proposed based on synchronizers, which collect state information from processes and threads, detect strongly consistent application global states, evaluate control predicates and send respective control signals. An algorithm for detection of strongly consistent global states for a variable number of processes/threads is presented.

Abstract: This paper presents an algorithm for trading resources in Grids. Resource description includes main technical attributes of a resource, such as processing power, memory capacity, etc., as well as a price. Trading is performed in a marketplace where providers' resources are matched with consumers' demand by means of auction mechanisms. The matching algorithm follows a strategy where a consumer's demand is matched with providers meeting the technical requirement and the price closest to the one offered by the consumer.

Abstract: A semantic enabled multi-agent architecture for solving non-linear equations systems by using a service oriented approach is proposed. The service oriented approach allows us to access already implemented methods for solving complex mathematical problems. The semantic descriptions of these services provide support for intelligent agents. The proposed architecture is a framework with two extension areas, agent society and domain ontology, with possible application in other domains too.

Abstract: Marketplaces for cloud services, referred to as electronic marketplaces (eMPs), have been introduced for a number of purposes, including comparability and transparency of service offerings. Since the complexity of services and their provisioning is increasing, eMPs need to handle the dynamic and automated creation of composite services, i.e. services which are yielded in part by using third-party services available on the marketplace. To meet this requirement, eMPs need a standardized machine readable description of service offerings. This paper presents an approach to define price models for services traded on eMPs, using a subset of the Unified Service Description Language (USDL). To enable composite service pricing, this paper introduces an approach on the aggregation of several price models into one price model and analyses the time complexity of the price aggregation algorithm presented.

Abstract: One of the main objectives of self-adaptive systems is to reduce maintenance costs through automatic adaptation. Self-healing is a self-adapting property that helps systems return to a normal state after a fault or vulnerability exploit has been detected. The problem is intuitively appealing as a way to automate the different type of maintenance processes (corrective, adaptive and perfective) and forms an interesting area of research that has inspired many initiatives. As a result, several surveys on self-healing have been published to describe the state of the art in this field. According to those surveys, the major trend towards finding a solution of the self-healing problem relies on redundancy that may concern both architecture and code resources. These approaches are therefore better suited to address adaptive and perfective maintenance. As part of the EU FP7 FastFix project, we focus on self-healing for corrective maintenance. We propose a framework for automating corrective maintenance that is based on software control principles. Our approach automates the engineering of self-healing systems as it does not require the system to be designed in a specific way. Instead it can be applied to legacy systems and automatically equip them with observation and control points. Moreover, the proposed approach relies on a sound control theory developed for Discrete Event Systems. Finally, this paper contributes to the field by introducing challenges to the effective application of this approach to relevant industrial systems. Some of these challenges are currently being tackled within FastFix.

Abstract: Worldwide accessibility of clouds brings great benefits by providing easy access to resources However, scheduling cloud resources for utilization among multiple collaborating cloud users is still often executed manually To address this problem, we developed a scheduling service for cloud middleware that guarantees optimal resource utilization in terms of a total number of used resources in a given interval based on user-defined policies In the paper, we introduce the scheduling algorithm, describe its supporting system architecture and provide the evaluation that proves the feasibility of the developed solution The provided scheduling algorithm takes into account dependencies between individual services, and can enforce common use of shared resources that lead to the optimal resource utilization By assuring continuous schedule optimality, costs caused by unnecessary usage of additional cloud resources are minimized

Abstract: The paper is concerned with a new distributed program design environment based on the global application states monitoring. The environment called PEGASUS (from Program Execution Governed by Asynchronous SUpervision of States) supplies to a programmer a ready to use control primitives to design distributed program execution control in which decisions for synchronous and asynchronous control actions are based on predicates evaluated on global application states. Such strongly consistent global application states are automatically constructed by the run-time system which additionally provides mechanisms for their analysis and organizing the respective program execution control in processes and threads of user programs executed in multicore processors. The PEGASUS control mechanisms are graphically supported in the respective program design framework. The paper first presents main general features of the PEGASUS environment. Next, it presents a method for load balancing inside distributed programs based on a set of parameters which are dynamically measured during program execution. Then, the paper presents how the described load balancing method can be implemented inside the PEGASUS environment taking as an example distributed programs for solving the Traveling Salesman Problem (TSP).

Abstract: In the paper we analyse results from the inversion of geophysical anomalies in high performance computing platforms. We experiment the solution of this ill-posed problem, trying to bypass the complexity of the calculations using simple algorithms that require huge calculation capacities offered by parallel systems. The gravity anomalies are considered because of the simplicity of the gravity modeling in geophysics.

Abstract: Additional communication links between unconnected agents are used in asynchronous searching, in order to detect obsolete information. A first way to remove obsolete information is to add new communication links, which allow a nogood owner to determine whether this nogood is obsolete or not. The second solution consists in temporarily keeping the links. A new link is maintained until a fixed number of messages have been exchanged through it. This article investigates different values for the number of messages, values that are either statically or dynamically, during the run time, determined. In the case of processing all the messages, we adapt a dynamical solution for determining the number of necessary messages for maintaining a connection. The experiments show a better efficiency in comparison with the standard Asynchronous Backtracking. In this paper we examine the effect of temporary links for the random binary constraints problem. Experiments with asynchronous search techniques are conducted on randomly generated networks of constraints. Experimental results show that the dynamical solution for the temporary links allows obtaining better results for the majority of classes of problems investigated.

Abstract: We develop and implement a ring-based parallel 3-D oil-phase homogeneous isotropic reservoir simulator and study its performance in terms of speedup as a function of problem size. The ring-based approach is shown to result in significant improvement in speedup as the problem size increases. This improvement stems from the reduction in communication costs inherent in a ring-based approach. The simulator employs a parallel conjugate gradient (CG) algorithm that we develop for solving the associated system of linear equations. The parallelization uses an MPI programming model. Previously proposed parallel oil reservoir simulators focus on data parallelism and load balancing and gives less attention to the communication cost. Performance analysis is given showing that the parallel algorithm results in a speedup of more than 42 times compared to a sequential simulator for a large simulation problem. This major improvement occurs for larger problem sizes, since the communication savings become significant. We compare our results to the implementation of the parallel oil reservoir simulator using the Portable Extensible Toolkit for Scientific Computation (PETSc). Oil reservoir simulators are used for forecasting reservoir potential before costly drilling, and are essential for improving oil recovery from existing fields, helping to maximize oil production. The speedup gained through the technique presented here can result in major savings of engineering time and more accurate reservoir management, and in turn higher oil production. Existing simulators suffer from limited performance due to the huge numerical operations involved. To cope with the issue, engineers usually reduce the size of the simulation model to get results in an acceptable timeframe, sacrificing accuracy of the predictions. This article describes the proposed ring-based algorithm for parallelization and development of a 3-D oil phase reservoir simulator. The work is a prelude to further planned research to develop an extended simulator that applies to three phases (oil, gas, and water) and to a heterogeneous and non-isotropic.

Abstract: The paper introduces the challenges in modern workflow management in distributed environments spanning multiple cluster, grid and cloud systems. Recent developments in cloud computing infrastructures are presented and are referring how clouds can be incorporated into distributed workflow management, aside from local and grid systems considered so far. Several challenges concerning workflow definition, optimisation and execution are considered. These range from configuration, integration of business and scientific services, data management, dynamic monitoring and tracking, reusable workflow patterns, semantic search and distributed execution of distributed services. Finally, the author recommends a solution to these challenges based on the BeesyCluster middleware for distributed management of services with static and dynamic rescheduling within a market of services.