Adaptive Reputation-Based Scheduling on Unreliable Distributed Infrastructures

This paper addresses the inherent unreliability and instability of worker nodes in large-scale donation-based distributed infrastructures such as peer-to-peer and grid systems. We present adaptive scheduling techniques that can mitigate this uncertainty and significantly outperform current approaches. In this work, we consider nodes that execute tasks via donated computational resources and may behave erratically or maliciously. We present a model in which reliability is not a binary property, but a statistical one based on a node's prior performance and behavior. We use this model to construct several reputation-based scheduling algorithms that employ estimated reliability ratings of worker nodes for efficient task allocation. Our scheduling algorithms are designed to adapt to changing system conditions, as well as nonstationary node reliability. Through simulation, we demonstrate that our algorithms can significantly improve throughput while maintaining a very high success rate of task completion. Our results suggest that reputation-based scheduling can handle a wide variety of worker populations, including nonstationary behavior, with overhead that scales well with system size. We also show that our adaptation mechanism allows the application designer fine-grain control over the desired performance metrics.     

Collective Practices in Common Information Spaces: Insight From Two Case Studies

This research explores the design of practice toolkits as components, distinct from community management systems, allowing members of a virtual community to engage in the practice the community is about. Our analysis is informed by two case studies in different application domains each presenting alternative but complementary insights to the design of computer-mediated practice toolkits. The first case study describes how established practices in music performance are encapsulated in a suitably augmented music notation toolkit so as to support the learning objectives of virtual teams engaged in music master classes. The second case study presents experience with the development of a toolkit for engaging in the practice of vacation package assembly. This time the virtual team is a cross-organization virtual community of practice whose members streamline their efforts by internalizing and performing in accordance to a new (virtual) practice. Findings from the two studies reveal two distinct orientations in the design of practice toolkits. Specifically, in application domains where practices are well established (i.e., music performance), the toolkit serves as the medium for reconstructing an existing practice in virtual settings. In contrast, when cross-organization collaboration is involved (i.e., vacation package assembly), the toolkit should be designed so as to encapsulate a “meta”-practice, exhibiting both boundary and locality.     

Modern Information Infrastructure in the Support of Distributed Collective Practice in Transport

Transport is one of the oldest and most important forms of distributed collective practice. This paper traces the role of information and communication technologies in the transformation of transport-based distributed collective practice, focusing on the evolution of technologies that place control of the transport infrastructure in the hands of end users. Examples of this shift are provided, including an analysis of the events of September 11, 2001 as forms of distributed collective action.     

Fine-Grain Interoperability of Scientific Workflows in Distributed Computing Infrastructures

Today there exist a wide variety of scientific workflow management systems, each designed to fulfill the needs of a certain scientific community. Unfortunately, once a workflow application has been designed in one particular system it becomes very hard to share it with users working with different systems. Portability of workflows and interoperability between current systems barely exists. In this work, we present the fine-grained interoperability solution proposed in the SHIWA European project that brings together four representative European workflow systems: ASKALON, MOTEUR, WS-PGRADE, and Triana. The proposed interoperability is realised at two levels of abstraction: abstract and concrete. At the abstract level, we propose a generic Interoperable Workflow Intermediate Representation (IWIR) that can be used as a common bridge for translating workflows between different languages independent of the underlying distributed computing infrastructure. At the concrete level, we propose a bundling technique that aggregates the abstract IWIR representation and concrete task representations to enable workflow instantiation, execution and scheduling. We illustrate case studies using two real-workflow applications designed in a native environment and then translated and executed by a foreign workflow system in a foreign distributed computing infrastructure.     

A Set of Successive Job Allocation Models in Distributed Computing Infrastructures

The growing number of scientific computation-intensive applications calls for an efficient utilization of large-scale, potentially interoperable distributed infrastructures. Parameter sweep applications represent a large body of workflows. While the principle of workflows is easy to conceive, their execution is very complex and no universally accepted solution exists. In this paper we focus on the resource allocation challenges of parameter study jobs in distributed computing infrastructures. To cope with this NP-hard problem and the high uncertainty present in these systems, we propose a series of job allocation models that helps refining and simplifying the problem complexity. In this way we present some special cases that are polynomial and show how more complex scenarios can be reduced to these models. It is known from practice that a small number of job sizes improves the result of job allocation, therefore we state a hypothesis relying on this fact in one of our models. Unfortunately, the reduction of the general problem (using K-means clustering) did not help, and thus the hypothesis has proved to be false. In the future, we shall look for clustering techniques which fit this goal better.     

Cost–Performance Optimization of Application- and Context-Aware Distributed Infrastructures

Application- and context-aware infrastructures involve the network in the execution of distributed applications through special devices, namely, the application cards, placed in network nodes. The sharp separation of applications and network is smoothed, and by performing part of the distributed application inside the network, it is possible to reduce costs and improve performance with a better optimization of the whole distributed information and communication technology (ICT) infrastructure. This optimization is allowed by the additional degrees of freedom of placing cards in nodes and of assigning applications to such cards. In this paper, we provide an optimization algorithm that minimizes the total cost of the entire distributed ICT infrastructure, given a target performance objective defined as the end-to-end delay for the completion of the distributed application tasks. We focus on two sample applications that are well suited for application- and context-aware infrastructures: caching and protocol translation. The joint optimization of computing and communication resources is an innovative contribution of this paper, as, in the literature, hardware and network components are typically optimized separately. Results show that the total infrastructural cost savings are in the range of 15%-20%. However, savings can be obtained only if cards satisfy a cost-performance curve that is also analyzed.     

为分布信息共享的网格信息服务

本文介绍了一个信息服务体系结构来解决网络的性能、安全、可扩展性和健壮性等问题。这个体系结构的定义包括支持不同查询语言与策略的聚合目录等低水平的查询和注册协议,容易将个人实体并入信息结构中。这些协议也能与其他网络协议一起构造高水平的服务和性能,例如:代理、监听、错误检测、故障修理。     

The Role of Integration in Health-Based Information Infrastructures

In this paper, we contribute with empirical insight into the complexity of establishing and sustaining integration between different information infrastructures in health care. An overall concern is to elaborate on how, despite many obstacles, the integration effort moves forward. We see this as a collective achievement, where users have an essential role in terms of mobilizing and coordinating the other actors as well as maintaining the integration. These activities are not limited to a specific project; they emerge from and are part of day-to-day practice. Empirically, we focus on a large integration initiative between the laboratory systems at the University Hospital of Northern Norway and the electronic patient records used by general practitioners in the Northern health region. Together with the vendor, Well Diagnostics, the hospital initiated a project aimed at establishing a new laboratory requisition system that enabled GPs to send requisitions electronically to the hospital laboratories. Theoretically, we draw on the concept of information infrastructures, and supplement this with Actor Network Theory.     

Collective Intentions and the Maintenance of Social Practices

This paper studies social practices and the dynamics of their maintenance in precise mathematical and logical terms. Roughly, social practices (such shaking hands, going regularly to sauna, eating a certain kind of food at Easter, etc.) are recurrent collective activities based on collective attitudes (shared we-attitudes), and the central kind of social practice under study here, is one based on collective intention. Social practices are the building blocks of human societies, and also of artificial ones. In this paper, we present a mathematical model representing the dynamics of social practices, and some properties of the process are subjected to detailed mathematical study.     

Scientific Data Collections and Distributed Collective Practice

As the basic sciences become increasingly information-intensive, the management and use of research data presents new challenges in the collective activities that constitute scholarly and scientific communication. This also presents new opportunities for understanding the role of informatics in scientific work practices, and for designing new kinds tools and resources needed to support them. These issues of data management, scientific communication and collective activity are brought together at once in scientific data collections (SDCs). What can the development and use of shared SDCs tell us about collective activity, dynamic infrastructures, and distributed scientific work? Using examples drawn from a nascent neuroscience data collection, we examine some unique features of SDCs to illustrate that they do more than act as infrastructures for scientific research. Instead, we argue that they are themselves instantiations of Distributed Collective Practice (DCP), and as such illustrate concepts of transition, emergence, and interdependency that may not be so apparent in other kinds of DCPs. We propose that research into SDCs can yield new insights into institutional arrangements, policymaking, and authority structures in other very large-scale socio-technical networks.     

Distributed Clustering Using Collective Principal Component Analysis

This paper considers distributed clustering of high-dimensional heterogeneous data using a distributed principal component analysis (PCA) technique called the collective PCA. It presents the collective PCA technique, which can be used independent of the clustering application. It shows a way to integrate the Collective PCA with a given off-the-shelf clustering algorithm in order to develop a distributed clustering technique. It also presents experimental results using different test data sets including an application for web mining. Received 30 August 2000 / Revised 30 January 2001 / Accepted in revised form 16 May 2001     

Efficient Remote Data Possession Checking in Critical Information Infrastructures

Checking data possession in networked information systems such as those related to critical infrastructures (power facilities, airports, data vaults, defense systems, etc.) is a matter of crucial importance. Remote data possession checking protocols permit to check that a remote server can access an uncorrupted file in such a way that the verifier does not need to know beforehand the entire file that is being verified. Unfortunately, current protocols only allow a limited number of successive verifications or are impractical from the computational point of view. In this paper, we present a new remote data possession checking protocol such that: 1) it allows an unlimited number of file integrity verifications; 2) its maximum running time can be chosen at set-up time and traded off against storage at the verifier.     

Distributed Multivariate Regression Using Wavelet-Based Collective Data Mining

This paper presents a method for distributed multivariate regression using wavelet-based collective data mining (CDM). The method seamlessly blends machine learning and the theory of communication with the statistical methods employed in parametric multivariate regression to provide an effective data mining technique for use in a distributed data and computation environment. The technique is applied to two benchmark data sets, producing results that are consistent with those obtained by applying standard parametric regression techniques to centralized data sets. Evaluation of the method in terms of mode accuracy as a function of appropriateness of the selected wavelet function, relative number of nonlinear cross-terms, and sample size demonstrates that accurate parametric multivariate regression models can be generated from distributed, heterogeneous, data sets with minimal data communication overhead compared to that required to aggregate a distributed data set. Application of this method to linear discriminant analysis, which is related to parametric multivariate regression, produced classification results on the Iris data set that are comparable to those obtained with centralized data analysis.     

Practices in Design Concept Selection as Distributed Cognition

A study was conducted of the practices that engineering designers had learned from experience to apply during the search for, and implementation of, new solution concepts. Each of 36 practices was analysed in terms of the goal it was directed to, and how it distributed cognition beyond the mental world of the designer. The results suggest that designers distribute cognition over their environments in a wide variety of ways which are not restricted, for instance, to using design tools. They also suggest that designers learn these practices in the context of specific experiences, probably by trial and error or social observation rather than means–end analysis. All but two of the practices involved distributed cognition of some kind, but there were no cases where this distribution involved ceding executive control to the environment.     

Collaborative Support for Informal Information in Collective Memory Systems

Informal information, such as the expertise of an organization or the workarounds practiced by a community, is a critical part of organizational or collective memory systems. From a user-centered perspective, a user merely wishes to get his work done, and to do this, he must solve his immediate problems. We have examined how to incorporate this problem solving into a collective memory, as well as how to incorporate the learning that accrues to it or from it. We report here on two systems, the Cafe ConstructionKit and the Collaborative Refinery, as well as an application, Answer Garden 2, built using these two systems. The Cafe ConstructionKit provides toolkit mechanisms for incorporating communication flows among people (as well as agents) into an organizational memory framework, and the Collaborative Refinery system provides mechanisms for distilling and refining the informal information obtained through these communication flows. The Answer Garden 2 application demonstrates the utility of these two underlying systems.     

Collective Mining of Bayesian Networks from Distributed Heterogeneous Data

We present a collective approach to learning a Bayesian network from distributed heterogeneous data. In this approach, we first learn a local Bayesian network at each site using the local data. Then each site identifies the observations that are most likely to be evidence of coupling between local and non-local variables and transmits a subset of these observations to a central site. Another Bayesian network is learnt at the central site using the data transmitted from the local site. The local and central Bayesian networks are combined to obtain a collective Bayesian network, which models the entire data. Experimental results and theoretical justification that demonstrate the feasibility of our approach are presented.16 November 2001     

支持产品设计的分布式集成信息系统

对支持设计的分布式集成信息系统的结构及其实现的关键问题进行了讨论,对系统中数据存储与传递作了详细的论述,实现了基于Internet的分布式集成信息系统中知识的表达、存储与获取,最后,给出了基于DIIS思想实现的原型系统,文中的工作为实现基于Internet快速知识获取、提高产品设计水平、加快开发速度提供了一套解决方案。     

Distributed Integrated Information System for Product Design

A new approach to construct Distributed Integrated Information System for product design(DIIS) is presented. Design knowledge of parts and products is expressed in STEP form, it is neutral and independent to commercial CAD platforms. After compiling by STEP Developer and being pre/post processed, the binary codes of data dictionary model can be sent into the end user's design platform directly and readily supports the product design.