Search

Particle physics experiments rely extensively on computing and data services, making e-infrastructure an integral part of the research collaboration. Constructing and operating distributed computing can however be challenging for a smaller-scale collaboration. The Light Dark Matter eXperiment (LDMX) is a planned small-scale accelerator-based experiment to search for dark matter in the sub-GeV mass region. Finalizing the design of the detector relies on Monte-Carlo simulation of expected physics processes. A distributed computing pilot project was proposed to better utilize available resources at the collaborating institutes, and to improve scalability and reproducibility. This paper outlines the chosen lightweight distributed solution, presenting requirements, the component integration steps, and the experiences using a pilot system for tests with large-scale simulations. The system leverages existing technologies wherever possible, minimizing the need for software development, and deploys only non-intrusive components at the participating sites. The pilot proved that integrating existing components can dramatically reduce the effort needed to build and operate a distributed e-infrastructure, making it attainable even for smaller research collaborations. [ABSTRACT FROM AUTHOR]

Thanks to its RDataFrame interface, ROOT now supports the execution of the same physics analysis code both on a single machine and on a cluster of distributed resources. In the latter scenario, it is common to read the input ROOT datasets over the network from remote storage systems, which often increases the time it takes for physicists to obtain their results. Storing the remote files much closer to where the computations will run can bring latency and execution time down. Such a solution can be improved further by caching only the actual portion of the dataset that will be processed on each machine in the cluster, reusing it in subsequent executions on the same input data. This paper shows the benefits of applying different means of caching input data in a distributed ROOT RDataFrame analysis. Two such mechanisms will be applied to this kind of workflow with different configurations, namely caching on the same nodes that process data or caching on a separate server. [ABSTRACT FROM AUTHOR]

DIRAC and Rucio are two standard pieces of software widely used in the HEP domain. DIRAC provides Workload and Data Management functionalities, among other things, while Rucio is a dedicated, advanced Distributed Data Management system. Many communities that already use DIRAC have expressed their interest in using DIRAC for Workload Management in combination with Rucio for Data Management. In this paper, we describe the integration of the Rucio File Catalog into DIRAC that was initially developed for the Belle II collaboration. [ABSTRACT FROM AUTHOR]

The ATLAS experiment's software production and distribution on the grid benefits from a semi-automated infrastructure that provides up-to-date information about software usability and availability through the CVMFS distribution service for all relevant systems. The software development process uses a Continuous Integration pipeline involving testing, validation, packaging and installation steps. For opportunistic sites that can not access CVMFS, containerized releases are needed. These standalone containers are currently created manually to support Monte-Carlo data production at such sites. In this paper we will describe an automated procedure for the containerization of ATLAS software releases in the existing software development infrastructure, its motivation, integration and testing in the distributed computing system. [ABSTRACT FROM AUTHOR]

Following the outbreak of the COVID–19 pandemic, the ATLAS experiment considered how it could most efficiently contribute using its distributed computing resources. After considering many suggestions, examining several potential projects and following the advice of the CERN COVID Task Force, it was decided to engage in the Folding@Home initiative, which provides payloads that perform protein folding simulations. This paper describes how ATLAS made a significant contribution to this project over the summer of 2020. [ABSTRACT FROM AUTHOR]

The CMS experiment at the CERN LHC (Large Hadron Collider) relies on a distributed computing infrastructure to process the multi-petabyte datasets where the collision and simulated data are stored. A scalable and reliable monitoring system is required to ensure efficient operation of the distributed computing services, and to provide a comprehensive set of measurements of the system performances. In this paper we present the full stack of CMS monitoring applications, partly based on the MONIT infrastructure, a suite of monitoring services provided by the CERN IT department. These are complemented by a set of applications developed over the last few years by CMS, leveraging open-source technologies that are industry-standards in the IT world, such as Kubernetes and Prometheus. We discuss how this choice helped the adoption of common monitoring solutions within the experiment, and increased the level of automation in the operation and deployment of our services. [ABSTRACT FROM AUTHOR]

The CERN ATLAS Experiment successfully uses a worldwide distributed computing Grid infrastructure to support its physics programme at the Large Hadron Collider (LHC). The Grid workflow system PanDA routinely manages up to 700,000 concurrently running production and analysis jobs to process simulation and detector data. In total more than 500 PB of data are distributed over more than 150 sites in the WLCG and handled by the ATLAS data management system Rucio. To prepare for the ever growing data rate in future LHC runs new developments are underway to embrace industry accepted protocols and technologies, and utilize opportunistic resources in a standard way. This paper reviews how the Google and Amazon Cloud computing services have been seamlessly integrated as a Grid site within PanDA and Rucio. Performance and brief cost evaluations will be discussed. Such setups could offer advanced Cloud tool-sets and provide added value for analysis facilities that are under discussions for LHC Run-4. [ABSTRACT FROM AUTHOR]

The use of computer clusters for computational sciences including computational physics is vital as it provides computing power to crunch big numbers at a faster rate. In compute intensive applications that requires high resolution such as Monte Carlo simulation, the use of computer clusters in a grid form that supplies computational power to any nodes within the grid that needs computing power, has now become a necessity. In this paper, we described how the clusters running on a specific application could use resources within the grid, to run the applications to speed up the computing process. [ABSTRACT FROM AUTHOR]

Information Centric Networks (ICN) is the new paradigm that envisages to shift the Internet away from its existing Point-to-Point architecture to a data centric, where communication is based on named hosts rather than the information stored on these hosts. Name Resolution is the center of attraction for ICN, where Named Data Objects (NDO) are used for identifying the information and guiding for routing or forwarding inside ICN. Recently, several researches use distributed NRS to overcome the problem of interest flooding, congestion and overloading. Yet the distribution of NRS is based on random distribution. How to distribute the NRS is still an important and challenging problem. In this work, we address the problem of distribution of NRS by proposing a new mechanism called Distributed Name Resolution System (DNRS), by considering the time of publishing the NDOs in the NRS. This mechanism partitions the network to distribute the workload among NRSs by increasing storage capacity. In addition, partitioning the network increases flexibility and scalability of NRS. We evaluate the effectiveness of our proposed mechanism, which achieves lesser end-to-end delay with more average throughputs compared to random distribution of NRS without disturbing the underlying routing or forwarding strategies. [ABSTRACT FROM AUTHOR]

The development of the smart cities is strongly linked to the Internet of Things (IoTs), where a lot of quotidian objects with connection capabilities interact in many applications over the network. One of the more relevant branches of the IoTs is the Internet of Vehicles (IoV), the next step in the Vehicular Ad hoc Networks (VANETS) as the number of vehicles connected to the IoTs increases. The main purpose of IoV is to combine the resources of many vehicles, things and networks to provide a robust communication capability. In this context the Vehicular Cloud applications take a great importance harnessing the wealthy resources contributed by groups of vehicles to enable innovative information services on the road. In this paper we present the architecture and the main procedures of a Vehicular Cloud Computing model on top of a virtualization layer and evaluate the adequacy to some Vehicular Cloud applications of special interest in smart cities of three different flavors of a new algorithm called VNIBR (Intersection-Based Routing on Virtual Nodes): reactive, proactive and encounter-based. [ABSTRACT FROM AUTHOR]

Given ever-increasing circuit complexities, recent trends are driving the requirement for distributed timing analysis (DTA) in electronic design automation (EDA) tools. However, DTA has received little research attention so far and remains a critical problem. In this paper, we introduce a DTA framework for large designs. Our framework supports (1) general design partitions in distributed file systems, (2) non-blocking IO with event-driven loop for effective communication and computation overlap, and (3) an efficient messaging interface between application and network layers. The effectiveness and scalability of our framework has been evaluated on large hierarchical industry designs over a cluster with hundreds of machines. [ABSTRACT FROM AUTHOR]

This paper is about the embedding of expert system into Distributed storage of models software package. Basic Distributed storage of models functions and used in this storage the Models translation tools (Multitranslator) was reviewed including their explanation and structure. Expert system application was proposed to improve functionality of Multitranslator and bring more automation into multi-language models translation process. The structure of expert system and it's interaction with Models translation tools were considered. Also the expert system work within of Distributed storage of models package was explained. To implement the expert system for Distributed storage of models the learning process was used. User-expert filled the knowledge base of expert system mostly but for exceptions we used intelligent editor to complete the learning process. Intelligent editor sends request to user to fulfill knowledge base, which requires only modeling knowledge. Application of expert system behaves as the replacing of expert so it has to solve the same tasks in same consequence. Practically it results in reduction of time costs while translating models and allows avoiding the necessity of constant user-expert support. Overall application of expert system for Distributed storage of models software package with Multitranslator excludes many cases of exceptions during process of multilanguage models translation that allows to automatize this process more. [ABSTRACT FROM AUTHOR]

Cloud computing allows use of computing power, storage space and data network, allowing access management. The use of cloud applications is becoming more popular, because of the savings that business realizes. The organizations have developed applications that sell all or part of their business activity. The problem with application moved to the cloud or to create a new application on the cloud platform is popular. The paper suggest an approach to create a cloud computing applications. [ABSTRACT FROM AUTHOR]

Tracking of ships' motion and monitoring of maritime traffic can be performed with the use of distributed Kalman Filtering. However, some of the local Kalman Filters which constitute distributed estimation schemes may be based on inaccurate models of the vessel's dynamics or kinematics and in such a case the aggregate state estimate provided by the distributed filter is unreliable. To treat this problem the paper proposes a statistical method of optimized performance for the validation of Fuzzy Kalman Filters used in ship tracking. This statistical validation test is capable of detecting the faulty local filter within the distributed estimation method, even in the case of small errors in the local model's parameters which do not exceed 1% of the associated nominal values. [ABSTRACT FROM AUTHOR]

The paper deals with the designing and verification of distributed logic controllers. The control system is initially modelled with Petri nets and formally verified against structural and behavioral properties with the application of the temporal logic and model checking technique. After that it is decomposed into separate sequential automata that are working concurrently. Each of them is re-verified and if the validation is successful, the system can be finally implemented. [ABSTRACT FROM AUTHOR]

Cloud brokers are intermediaries supporting the evaluation, selection, and implementation of cloud services. Especially small and medium-sized enterprises with limited technical competencies struggle with a complex and non-transparent market, selecting and combining services from different cloud service providers. They need guidance in the pre-adoption phase to successfully handle selection decisions. Using decisional guidance theory, we have designed a cloud broker platform that aims to provide such guidance. The results of its evaluation show how such a platform improves selection decisions by reducing complexity and structuring the selection process. They also show that the platform’s usefulness may be limited in the case of very complex cloud services depending, for example, on the degree of customization needs. Our study provides design knowledge for decisional guidance (in the context of cloud broker platforms) and shows how cloud service selection can be implemented in practice. [ABSTRACT FROM AUTHOR]

The Belle II experiment started taking physics data in April 2018 with an estimated total volume of all files including raw events, Monte-Carlo and skim statistics of 340 petabytes expected by the end of operations in the late-2020s. Originally designed as a fully integrated component of the BelleDIRAC production system, the Belle II distributed data management (DDM) software needs to manage data across about 29 storage elements worldwide for a collaboration of nearly 1000 physicists. By late 2018, this software required significant performance improvements to meet the requirements of physics data taking and was seriously lacking in automation. Rucio, the DDM solution created by ATLAS, was an obvious alternative but required tight integration with BelleDIRAC and a seamless yet non-trivial migration. This contribution describes the work done on both DDM options, the current status of the software running successfully in production and the problems associated with trying to balance long-term operations cost against short term risk. [ABSTRACT FROM AUTHOR]

Efficient access to distributed computing and storage resources is mandatory for the success of current and future High Energy and Nuclear Physics Experiments. DIRAC is an interware to build and operate distributed computing systems. It provides a development framework and a rich set of services for the Workload, Data and Production Management tasks of large scientific communities. A single DIRAC installation provides a complete solution for the distributed computing of one, or more than one collaboration. The DIRAC Workload Management System (WMS) provides a transparent, uniform interface for managing computing resources. The DIRAC Data Management System (DMS) offers all the necessary tools to ensure data handling operations: it supports transparent access to storage resources based on multiple technologies, and is easily expandable. Distributed Data management can be performed, also using third party services, and operations are resilient with respect to failures. DIRAC is highly customizable and can be easily extended. For these reasons, a vast and heterogeneous set of scientific collaborations have adopted DIRAC as the base for their computing models. Users from different experiments can interact with the system in different ways, depending on their specific tasks, expertise level and previous experience using command line tools, python APIs or Web Portals. The requirements of the diverse DIRAC user communities and hosting infrastructures triggered multiple developments to improve the system usability: examples include the adoption of industry standard authorization and authentication infrastructure solutions, the management of diverse computing resources (cloud, HPC, GPGPU, etc.), the handling of high-intensity work and data flows, but also advanced monitoring and accounting using no-SQL based solutions and message queues. This contribution will highlight DIRAC's current, upcoming and planned capabilities and technologies. [ABSTRACT FROM AUTHOR]

Service-orientation for the development of complex distributed systems with reusable components (services) has proved to be highly beneficial approach in software engineering. In order to realize its advantages automated techniques and tools for finding services, selecting the ones that match requestor needs, composing them for achieving enhanced functionality and resolving heterogeneity issues are required. Adding semantics to web services has become the key approach for the automation of service-orientation environments. Current paper proposes semantic approach for service selection and integration. It considers the semantic description of web services and involves reasoning on a knowledge base. The proposed approach for web service discovery and integration is based on designing logic based techniques for matching goal description with the semantic annotation of web services. It provides for the automation of web service search which facilitates the design of practical applications in service-oriented environment. [ABSTRACT FROM AUTHOR]

High availability in complex distributed systems is a challenge that has to be considered not only at design time, but also at execution time. This paper proposes an automatized monitoring architecture, event driven, based in rules with the goal of minimizing systems unavailability. Translating the expertise that operators have to rules, and providing an appropriate interface for services operation, allow us to execute self-healing actions to anticipate or correct fault cases as quickly as possible. [ABSTRACT FROM AUTHOR]

We document ongoing research on a digital artefact comprising a settlement layer for financial contracts traded over the counter (OTC). Working with a team of stakeholders from a hosting organization we employ the design science research methodology (DSR) in the design, implementation and evaluation of a DLT based settlement layer for OTC contracts. Our findings indicate that DLT may serve towards reducing execution costs while mitigating counterparty credit risk through deterministic automation of key processes in the OTC contract lifecycle. Extrapolating these early results, we introduce three general propositions on the systemic and commercial implications of DLT. We contribute to the broader IS discourse on the capacity for emerging digital technologies to upend traditional knowledge processes in the financial services. [ABSTRACT FROM AUTHOR]

Debugging of distributed computing model programs like MapReduce is a difficult task. That's why prior studies only focus on finding and fixing bugs in early stages of program development. Delta debugging tries to find minimal failing input in sequential programs by dividing inputs into subsets and testing these subsets one-by-one. But no prior work tries to find minimal failing input in distributed programs like MapReduce. In this paper, we present MapRedDD, a framework to efficiently find minimal failing input in MapReduce programs. MapRedDD employs failing input selection technique, focused on identifying the failing input subset in the single run of MapReduce program with multiple input subsets instead of testing each subset separately. This helps to reduce the number of executions of MapReduce program for each input subset and overcome the overhead of job submission, job scheduling and final outcome retrieval. Our work can efficiently find the minimal failing input in the number of executions equal to the number of inputs to MapReduce program N as opposed to the number of executions of MapReduce program equal to the number of input subsets 2N - 1 in worst case for binary search invariant algorithm to find minimal failing input. [ABSTRACT FROM AUTHOR]

Symbolic execution is a systematic program analysis technique that has received a lot of attention in the research community. However, scaling symbolic execution continues to pose a major challenge. This paper introduces SynergiSE, a novel two-fold integration approach. One, it integrates distributed analysis and constraint re-use to enhance symbolic execution using feasible ranges, which allow sharing of constraint solving results among different workers without communicating or sharing potentially large constraint databases (as required traditionally). Two, it integrates complementary techniques for test input generation, e.g., search-based generation and symbolic execution, for creating higher quality tests using unexplored ranges, which allows symbolic execution to re-use tests created by another technique for effective distribution of exploration of previously unexplored paths. [ABSTRACT FROM AUTHOR]

In this paper, we propose a new software design of an online judge system for interactive theorem proving. The distinctive feature of this architecture is that our online judge system is distributed on the network and especially involves volunteer computing. In volunteers' computers, network bots (software robots) are executed and donate computational resources to the central host of the online judge system. Our proposed design improves fault tolerance and security. We gave an implementation to two different styles of interactive theorem prover, Coq and ACL2, and evaluated our proposed architecture. From the experiment on the implementation, we concluded that our architecture is efficient enough to be used practically. [ABSTRACT FROM AUTHOR]

This paper presents a pilot project that uses cloud-based documents to provide bi-directional feedback on open-ended contextualized activities. The course setting is a first year engineering design and professional practice course of approximately 700 students, taught in three sections, at Queen's University, a medium-sized research-intensive institution in Canada. Students were assigned to groups, and assigned shared directories in Google Drive. In most lectures students were assigned an activity requiring a response that was completed in a document in Google Drive, e.g. open-ended design problems, brainstorming, evaluation of information sources, etc. The course instructor, and a teaching assistant (TA) who attended each class, were able to view samples of students work and select some to anonymously discuss with the class. Thorough well-thought responses were highlighted, and misconceptions or misdirection was addressed. In this way students received some feedback on common issues in constructed response tasks immediately, rather than waiting for submission and grading of assignments. Student response to this approach was mixed, with about half the class feeling that the approach helped them learn, and that feedback was useful. About a third were neutral to the idea, and the rest felt it was not useful. There was no correlation between students' use of Google Docs and their grades on the two major reports. The instructor felt that this is a useful approach that helped many students, and would be improved in the future by using classrooms designed for team activities, improving the quality of tasks assigned in class, and assigning completion grades for in-class activities. [ABSTRACT FROM AUTHOR]

A professional development program for high-school teachers, designed to explore ways in which Cloud Computing technologies can be leveraged to improve classroom instruction, has been developed to support the educational component of the NSF CAREER grant awarded in 2011 to Dr. Reza Curtmola at the New Jersey Institute of Technology. The goal of the program was twofold: first, to expose high school teachers to the concept of Cloud Computing and the technologies associated with it, and second, to help teachers develop curriculum units based on Cloud Computing technologies that can be integrated into different high-school subjects. Sixteen high-school teachers participated in the program. The participants' teaching areas spanned a wide array of subjects ranging from Computer Technology, Math, Physics and Chemistry, to Human Anatomy, Biology, Environmental Science, and even Spanish. Participants received 20 hours of professional development credit. The program was structured into two workshop sessions. The first session, during the summer of 2012, consisted of three days of hands-on instruction. It focused on several instructional topics, including (a) Overview of Cloud Services, (b) Storing and Sharing Data in the Cloud, (c) Clouds in Education and Collaboration in and out of the Classroom, (d) Cloud-based Tools for Real-time Collaboration, (e) Course Management using Piazza, (f) Standards-based Lesson Planning and Post-workshop Assignment, (g) Creating a Lesson Plan, and (h) Using Public Data Sets Available in Amazon's Cloud. At the end of the three-day workshop, participants were given an assignment to identify a lesson taught in their classroom which could use the cloud as an educational technology tool and then to write a revised lesson plan based on cloud computing integration and standards-based lesson planning. The assignment also required participants to submit the revised and original lesson plans plus samples of student work. The submitted lesson plans reflected the diversity of subjects taught by the participants and shared with their peers. The completed assignments were presented at the second workshop session, which consisted of a 1-day meeting in December 2012. Based on the teachers' applications for the program, we were able to outline what the teachers hoped to learn from the workshops. A pre-workshop survey helped us assess the participants' knowledge of cloud computing before beginning the program. Two-post workshop surveys were used to evaluate the impact of the program: a first post-workshop survey at the end of the first session, and a second post-workshop survey at the end of the second session. This paper describes the instructional material used during the workshop, the group discussions during the second workshop session, the assignments and lessons plans completed by teachers, and an assessment of the workshop's impact based on an analysis of the three surveys. A follow- up survey about their classroom implementation experiences and subsequent lesson plans was sent to all participants at the end of the 2012-2013 academic year. Participants' responses provided useful feedback for future workshops. [ABSTRACT FROM AUTHOR]

In this article are considered the four classical principles of transactional approach in client/server systems. We define as "Classical" client-server systems the ones that are developed specifically for one company. ACID (Atomicity, Concurrency, Independence and Durability) are the four classical principles which are discussed in the paper. Classical client-server systems we are considering as non-distributed and distributed. The four classical principles has applied in a different way according to the system - non-distributed or distributed. [ABSTRACT FROM AUTHOR]

Pervasive computing is an important paradigm for interacting with the increasingly complex environment surrounding the user while social networking is equally important for handling the daily interactions between users. One challenge that has been identified recently is to bring these two paradigms together in an integrated way so that the user has the advantages of both, together with benefits arising from the combination of the two. This is the aim of the SOCIETIES project which is developing a system that combines general pervasive system behaviour with social networking in a seamless fashion. One of the important features on which the system is based, is that of context aware personalization, and major components in the SOCIETIES system include those of personalization and context management. Within this, one of the major challenges is that of building up a set of user preferences to obtain the best predictions, especially since users will change their minds about preferences from time to time. To achieve this two different approaches (rule-based and neural net) are used by the system to represent user preferences. This paper describes the smart space approach developed in the Persist project and currently being expanded in SOCIETIES, and then focuses on the problem of managing user preferences. The rule-based approach is described in detail and the neural net approach only briefly. The strategy used employs both approaches and compares the predictions of the two to obtain the best accuracy and adaptability. [ABSTRACT FROM AUTHOR]

Quadratic Assignment Problems (QAP) is classified as the NP hard problem. It has been used to model a lot of problem in several areas such as operational research, combinatorial data analysis and also parallel and distributed computing, optimization problem such as graph portioning and Travel Salesman Problem (TSP). In the literature, researcher use exact algorithm, heuristics algorithm and metaheuristic approaches to solve QAP problem. QAP is largely applied in facility layout problem (FLP). In this paper we used QAP to model university facility layout problem. There are 8 facilities that need to be assigned to 8 locations. Hence we have modeled a QAP problem with n ≤ 10 and developed an Ant Colony Optimization (ACO) algorithm to solve the university facility layout problem. The objective is to assign n facilities to n locations such that the minimum product of flows and distances is obtained. Flow is the movement from one to another facility, whereas distance is the distance between one locations of a facility to other facilities locations. The objective of the QAP is to obtain minimum total walking (flow) of lecturers from one destination to another (distance). [ABSTRACT FROM AUTHOR]

Both academia and industry have been putting a lot of effort into investigating various solutions to the problem of autonomic management of complex distributed computer systems. Existing approaches, based on the principles of autonomic computing, typically rely on monitoring a managed system, analysing the monitored values, and executing corresponding adaptation actions. By addressing existing limitations, we present a novel research area - stream reasoning - in the context of run-time monitoring and analysis in autonomic systems. This research area aims at enhancing existing stream processing techniques by adding dynamic reasoning support, thereby fitting scenarios when timely, yet precise analysis of dynamically flowing data is needed. In this paper, by discussing benefits and shortcomings, we speculate on the potential role of the approach in autonomic computer systems. [ABSTRACT FROM AUTHOR]

Significant changes are currently taking place in scientific research. In particular, there is a greater collaboration among large groups of researchers, which leads to an increase use of information processing techniques, and a need to share results and observations among the participants in the process. Using technologies such as peer-to-peer networks and semantic web, we create an access point to distributed scientific applications, where scientists can work with heterogeneous information and create scientific communities according to their specialties and interests. This paper presents the e-ScienceNet, which is an architecture that supports the creation of scientific communities in the context of semantic e-Science, facilitating the creation and sharing of scientific data and services. This paper also presents the e- ScienceNet prototype. [ABSTRACT FROM AUTHOR]

This paper proposes a model of a colluding attack in Volunteer Computing (VC), where some of participants are unreliable and may return incorrect results. The colluding attack is one of the most important issues to realize reliable VC because it may ruin the most basic assumption, i.e. "the majority in voting is correct". Especially, check-by-voting archives the largest efficiency by sorting reliable participants based on whether their results are the majority or not. Thus, if some incorrect results become the majority by colluding attack, it may have significant impacts on check-by-voting. In this paper, we perform a Monte Carlo simulation of VC using the proposed colluding model and evaluate the sabotage-tolerance performance of voting methods. Simulation results show that check-by-voting works well if colluding attack happens. [ABSTRACT FROM AUTHOR]

Distributed systems often contain many behaviorally similar processes, which are conveniently grouped into classes. In system modeling, it is common to specify such systems by describing the class level behavior, instead of object level behavior. While there have been techniques that mine specifications of such distributed systems from their execution traces, these methods only mine object-level specifications involving concrete process objects. This leads to specifications which are large, hard to comprehend, and sensitive to simple changes in the system (such as the number of objects). In this paper, we develop a class level specification mining framework for distributed systems. A specification that describes interaction snippets between various processes in a distributed system forms a natural and intuitive way to document their behavior. Our mining method groups together such interactions between behaviorally similar processes, and presents a mined specification involving "symbolic" Message Sequence Charts. Our experiments indicate that our mined symbolic specifications are significantly smaller than mined concrete specifications, while at the same time achieving better precision and recall. [ABSTRACT FROM AUTHOR]

In this paper we consider the core issues that are inhibiting the application of general Service Oriented Architecture (SOA) principles to distributed real-time military environments. In particular, we propose a Distributed Real-Time Service Oriented Architecture (DRT-SOA) framework intended to overcome some of the problems that distributed SOA systems face in needing to exhibit verifiable deterministic behaviour. Although the time behaviour of asynchronous systems is inherently non-deterministic, such systems offer advantages in terms of extensibility, cost and utilisation. We show that DRT-SOA presents a framework where the sources of non-determinism can be supressed, making it suitable for certain classes of real-time system. We also show that decentralised execution of complex workflows is achievable using dynamic rule-based logic. We also present preliminary results demonstrating the architecture's ability to perform decentralised execution of workflow and to resolve deadline constrained tasks. We also present results of a sensitivity analysis showing that the architectural framework is able to scale in support of dynamic redeployment of workflow. This has application when needing to adapt to changing run-time needs in the case of self-healing. [ABSTRACT FROM PUBLISHER]

Many complex systems, such as multi-agent modeling and simulation which are crucial in military applications, necessitate concurrent task planning and decision making. These applications include battlefield modeling and simulation for soldier virtual training. In general these systems are implemented on automated systems with distributed architectures on heterogeneous operating systems and hardware platforms. In this paper we present an autonomous multi agent path planning system with agents-in-formation constraints in a distributed computing environment. Here a given agent, an entity with behaviors, exchanges sensory positioning feedback information over the network with other agents that are running on separate networked distributed computing nodes than the one the given agent is running on. Hence the agents do not have complete knowledge of the environment, but only a limited one. That is, in keeping with the agent formation constraint, the agents have knowledge only of the agents' information residing on the distributed memory of the computing node the agents are running on. The agents exchange agents' positioning sensory information of the rest of the agents they are interacting with, over a distributed network. Here agents autonomous trajectory planning is achieved for each agent to reach its destination via an unstructured distributed environment filled with dynamic obstacles and other agents. One focus is on the agents achieving their goals under real time formation constraints. We propose a multi-threaded architecture to achieve light-weight concurrency and multi-agent cooperative concurrency, as they allow separation between address space and thread control. Additionally, a threads-based architecture benefits from non-blocking system calls. Finally, experimental results are presented to show the potential of the proposed method for multi-agent distributed path planning. [ABSTRACT FROM PUBLISHER]

This paper describes the work carried out to implement a wireless sensor network capable of automatically evacuating a building. A distributed computing cloud was created in nesC for TinyOS and two different types of evacuee detection sensors were designed and tested. LED display boards were designed from first principles and used to guide evacuees to the nearest exit. Control room software was developed to display evacuation data to a control room operator and used a building overlay together with graphs and charts to display the information in several meaningful ways. The system was successfully deployed at the Council of Scientific and Industrial Research (CSIR) and several tests were conducted. The distributed computing cloud functioned as expected, directing evacuees along the most effective routes, based on routing metrics acquired during the evacuation. The system was able to re-route evacuees based on congestion estimates, additional evacuation triggers and manual overrides. [ABSTRACT FROM PUBLISHER]

We consider the problem of reliably broadcasting information in a multi hop asynchronous network that is subject to Byzantine failures. That is, some nodes of the network can exhibit arbitrary (and potentially malicious) behavior. Existing solutions provide deterministic guarantees for broadcasting between all correct nodes, but require that the communication network is highly-connected (typically, 2k+1 connectivity is required, where k is the total number of Byzantine nodes in the network). In this paper, we investigate the possibility of Byzantine tolerant reliable broadcast between most correct nodes in low-connectivity networks (typically, networks with constant connectivity). In more details, we propose a new broadcast protocol that is specifically designed for low-connectivity networks. We provide sufficient conditions for correct nodes using our protocol to reliably communicate despite Byzantine participants. We present experimental results that show that our approach is especially effective in low-connectivity networks when Byzantine nodes are randomly distributed. [ABSTRACT FROM PUBLISHER]

High-performance scientific applications require more and more compute power. The concurrent use of multiple distributed compute resources is vital for making scientific progress. The resulting distributed system, a so-called Jungle Computing System, is both highly heterogeneous and hierarchical, potentially consisting of grids, clouds, stand-alone machines, clusters, desktop grids, mobile devices, and supercomputers, possibly with accelerators such as GPUs. One striking example of applications that can benefit greatly of Jungle Computing Systems are Multi-Model / Multi-Kernel simulations. In these simulations, multiple models, possibly implemented using different techniques and programming models, are coupled into a single simulation of a physical system. Examples include the domain of computational astrophysics and climate modeling. In this paper we investigate the use of Jungle Computing Systems for such Multi-Model / Multi-Kernel simulations. We make use of the software developed in the Ibis project, which addresses many of the problems faced when running applications on Jungle Computing Systems. We create a prototype Jungle-aware version of AMUSE, an astrophysical simulation framework. We show preliminary experiments with the resulting system, using clusters, grids, stand-alone machines, and GPUs. [ABSTRACT FROM PUBLISHER]

Small-world networks have received significant attention because of their potential as models for the interaction networks of complex systems. Specifically, neither random networks nor regular lattices seem to be an adequate framework within which to study real-world complex systems such as chemical-reaction networks, neural networks, food webs, social networks, scientific-collaboration networks, and computer networks. Small-world networks provide some desired properties like an expected poly logarithmic distance between two processes in the network, which allows routing in poly logarithmic hops by simple greedy routing, and robustness against attacks or failures. By these properties, small-world networks are possible solutions for large overlay networks comparable to structured overlay networks like CAN, Pastry, Chord, which also provide poly logarithmic routing, but due to their uniform structure, structured overlay networks are more vulnerable to attacks or failures. In this paper we bring together a randomized process converging to a small-world network and a self-stabilization process so that a small-world network is formed out of any weakly connected initial state. To the best of our knowledge this is the first distributed self-stabilization process for building a small-world network. [ABSTRACT FROM PUBLISHER]

Online anomaly detection is critical for protecting wireless sensor networks (WSNs) from cyber-attacks and random faults, which handles the streaming data in real-time. Comparing to other techniques, histogram-based anomaly detection is cheaper in computation, which should be suitable for WSNs. However, performing histogram-based anomaly detection with an online manner in WSNs is not a straightforward issue. Most of the existing histogram-based schemes have to depend on a verification procedure, which costs a great amount of computational overhead as well as communication overhead. Thus, it almost wipes out the advantage of low complexity of histogram-based anomaly detection. This paper introduces a simple estimating approach to detect anomalies with the histogram, which takes account into the distributed manner and online manner at the same time. It also proves the error caused by the new estimate is very small, through a theoretical analysis. Moreover, the optimal parameter will be suggested by minimizing the error. Finally, a set of experiments are implemented with a real WSN dataset, which prove the new scheme is effective and efficient. [ABSTRACT FROM PUBLISHER]

In this paper, we have proposed a simple and efficient approach for recovery in distributed computing environment. The check pointing scheme used in the work ensures that after the system recovers from failures, all processes can restart from their respective recent checkpoints; thus avoiding any domino effect. That is, the recent check points always form a consistent recovery line. The recovery scheme deals with both orphan and lost messages. Therefore, correctness of computation of the underlying application program is guaranteed. [ABSTRACT FROM PUBLISHER]

The use of computer systems have been widely used in such applications, scientific, manufacturing and business. The main idea of distributed computing systems is to use a client-server architecture in combination with remote method invocation. The invocation of these methods is to send data from a client to a server for processing and then the server returns data to the client for presentation. In general, distributed platforms Java-RMI, CORBA and Web Services provide a range of methods to communicate with distributed objects on remote computers. This paper presents a comparative study (measurement of processor usage, memory, packets on the network, etc.) Between CORBA and Web Services, implementing two distributed services, a basic arithmetic operations and sending strings. [ABSTRACT FROM PUBLISHER]

MapReduce is a common programming model for processing and generating large datasets at present. Using the model, the programming for distributed computing can be easier than others. The extremal graph is a graph with the maximum number of edges such that it does not contain given sub graph. The method of constructing extremal graphs is an important research content in Graph Theory. The algorithm of constructing extremal graphs based on the MapReduce programming model is studied in this paper. By mapping the key-value pairs of MapReduce model properly, the parallel constructing algorithms are designed and implemented. Finally, we construct extremal graphs not containing hexagons and with no more than 26 vertices using the algorithm. By the results of experiments, it is showed that the average speedup is 2.55, and the average efficiency is 85%. [ABSTRACT FROM PUBLISHER]

In the era of Internet, one of the most ultimate goals is that users can seamlessly access to any computing application or service at anywhere anytime. Computing infrastructuralization is the path to gradually address the challenge. In this work, elop computing is proposed that encapsulates and integrates various computing elements as services, benefiting from existing advanced computing technology, e.g. virtualization and SaaS. Higher levels of services are also provided, e.g. metadata management, resource management and scheduling, security and authorization, which can be utilized to build distributed computing systems in a scalable way. In this paper, the architecture of elop computing is illustrated and corresponding middleware implementation is introduced in details. A typical scenario is given to demonstrate the potential of enabling distributed computing systems with elopTM. [ABSTRACT FROM PUBLISHER]

In order to improve the quality of teaching, this paper presents a model of a negative feedback process of teaching reform. The model can be developed to improve the teaching program to develop students' ability. The paper discusses the system of two types of indicators to measure teaching effectiveness, as well as a corresponding meaning. Also we propose how to conduct data fusion algorithms and architecture model. Finally, the experimental data is discussed for the effect of the teaching quality improvement for teaching reform. [ABSTRACT FROM PUBLISHER]

Lazy controllers are a class of execution monitors that do not continuously observe the behaviour of their target. Monitors are activated and deactivated according to a scheduling strategy. When a lazy controller is activated, it checks the current security state and, in case of a violation, terminates the execution. Instead, if the current execution trace is safe, the monitor is suspended and its activation is scheduled again. The inactivity period is computed by considering the risk that, from the current state, the target can reach a faulty configuration. This behaviour is particularly interesting for systems which is difficult to monitor with standard solutions, e.g., web services. In this paper we present a prototype using existing logging API, i.e., the Commons Logging Package, for remotely watching the execution of OSGi bundles. We claim that our solution can efficiently follow the target system keeping under control the delay in detecting violations. Also, as we use standard OSGi platform and facilities, we show that our monitors can run under very realistic assumptions in the context of web services. [ABSTRACT FROM PUBLISHER]

Peer-to-Peer(P2P) computing is a new approach to establish a high performance computing system which aggregates idle computing cycles of PCs available on Internet. In this paper we propose a general purpose pure P2P computing framework for large scale scientific computation problems. Peer groups are fundamental building blocks of proposed framework. We propose four peer groups-administrator, query manager, task distributor and task processor group, and also define role and functionalities of these peer groups in framework. The framework provides a conceptual blueprint for a pure P2P computing system which is useful for general users to process their large computation problems. [ABSTRACT FROM PUBLISHER]

The widely adopted and implemented core web services standards (SOAP and WSDL)[10] have achieved extraordinary interoperability across highly disparate software systems. The service oriented architecture (SOA)[4] has become widely recognized for its important role in information technology (IT) projects. A SOA is a style of design that guides an organization during all aspects of creating and using business services (including conception, modelling, design, development, deployment and management). SOA has been the ideal combination of architecture and technology for consistently delivering robust, reusable services that support today's business needs and that can be easily adapted to satisfy changing business requirements. As systems become more complex, the overall system structure-or architecture- --becomes a central design problem. A system's architecture provides a model of the system that suppresses implementation detail. Unfortunately, current representations of SOA architecture are informal and ad hoc. Currently many state of the art formal methods have been applied into the modelling, interoperability, dependability and trustworthiness of web services and this could have a significant impact on the ongoing standardization efforts for services and cloud technologies. This paper presents a formal verification of proposed x-SOA based architecture for UDDI based web service discovery framework. Potentially, extending the state of art formal method techniques could have a significant impact on the ongoing standardization efforts for web services and cloud technologies. [ABSTRACT FROM AUTHOR]

In this paper, we have highlighted the significance of open source software for distributed computing environment. We have identified the open source software available for distributed computing environment and proposed a need of incorporation of dynamic load balancing feature in the commonly used Open source software. We have also proposed a model of a load balancing system useful for the open source software development. [ABSTRACT FROM PUBLISHER]

OddCI is a new architecture for distributed computing that is, at same time, flexible and highly scalable. Previous works have demonstrated the theoretical feasibility of implementing the proposed architecture on a digital television (DTV) network, but without taking into consideration any practical issues or details. This paper describes the implementation of a proof of concept for the architecture, called OddCI-Ginga, using a testbed based on DTV receivers compatible with the Brazilian DTV System. Performance tests using real broadcast transmission and the return channel demonstrate the feasibility of the model and its usefulness as a platform for efficient and scalable distributed computing. [ABSTRACT FROM PUBLISHER]