Network function virtualization is a promising technique to greatly improve the effectiveness and flexibility of network services through a process named service function chain (SFC) mapping, with which network functions are deployed over virtualized and shared platforms in data centers. However, failures are quite common in data centers. Therefore, a practical and yet theoretically challenging issue in SFC mapping in such an environment is to manage the availability of the requests. In this paper, we present a framework to provision availability of SFC requests in a data center with multiple layers of connected devices, and the devices follow heterogeneous failure processes with the objective of minimizing resource usage. To expedite the process, we further propose an optimization problem of request mapping and backup estimation and solve it efficiently with an approximation algorithm. With simulations, we demonstrate the effectiveness of our proposed framework.

Availability, as a part of Service Level Agreement (SLA), is a critically important issue in cloud services, as an application may not be able to run after certain server or network failures. Cloud service providers seek to not only fulfill the SLA, but also simultaneously minimize their operating costs, which are dominated by the energy consumption. In order to minimize the impact of a server/switch failure inside the datacenter on a single application, one would like to spread out the Virtual Machines (VM) for the application across different racks. However, in doing so, the power consumption may increase significantly. In this paper, we develop a variance-based metric to measure the risk of violating the availability requirement. We then propose two heuristic algorithms to place VMs in online and offline manners, respectively. These algorithms aim to strike a balance between minimizing the risk of violating the availability requirement and minimizing the energy, in order to reduce the overall cost.

OBJECTIVE: The objective of this research is to empirically explore the drivers of patients' consent to sharing of their medical records on health information exchange (HIE) platforms. MATERIALS AND METHODS: The authors analyze a dataset consisting of consent choices of 20,076 patients in Western New York. A logistic regression is applied to empirically investigate the effects of patients' age, gender, complexity of medical conditions, and the role of primary care physicians on patients' willingness to disclose medical information on HIE platforms. RESULTS: The likelihood of providing consent increases by age (odds ratio (OR) = 1.055; P < .0001). Female patients are more likely to provide consent (OR = 1.460; P = .0003). As the number of different physicians involved in the care of the patient increases, the odds of providing consent slightly increases (OR = 1.024; P = .0031). The odds of providing consent is significantly higher for the patients whom a primary care physician has been involved in their medical care (OR = 1.323; P < .0001). CONCLUSION: Individual-level characteristics are important predictors of patients' willingness to disclose their medical information on HIE platforms.

A central challenge in developing cloud datacenters Service Level Agreements is the estimation of downtime distribution of a set of provisioned servers over a service window, which is compounded by three facts. First, while steady-state probabilities have been derived for birth-death processes involving server failures and repairs, they could be highly inaccurate under transience. Furthermore, steady-state cannot be assured under typical service windows. Therefore, estimation of transient distributions is essential. Second, the processes of failures and repairs may follow any distribution and hence need to be extracted using system log data and modeled using appropriate general distributions. Third, downtime distributions over service windows depend on the number of servers and their deployment structure for a contract. We develop an efficient and generalized sample path randomization approach to precisely estimate transient probabilities under three different checkpointing strategies and three flexible failure distribution models. The estimators are unbiased, consistent, efficient and sufficient. Their asymptotic convergence is established. The estimation algorithms are computationally efficient in solving practical problems and yield rich information on transient system behaviors. The methodology is general and extensible to various server failure and repair processes characterized using birth-death modeling.

BACKGROUND AND OBJECTIVE: We postulate that professional proximity due to common patients and geographical proximity among practice locations are significant factors influencing the adoption of health information exchange (HIE) services by healthcare providers. The objective of this study is to investigate the direct and indirect network effects of these drivers on HIE diffusion. DESIGN: Multi-dimensional scaling and clustering are first used to create different clusters of physicians based on their professional and geographical proximities. Extending the Bass diffusion model to capture direct and indirect network effects among groups, the growth of HIE among these clusters is modeled and studied. The network effects among the clusters are investigated using adoption data over a 3-year period for an HIE based in Western New York. MEASUREMENT: HIE adoption parameters-external sources of influence as well as direct and indirect network coefficients-are estimated by the extended version of the Bass diffusion model. RESULTS: Direct network effects caused by common patients among physicians are much more influential on HIE adoption as compared with previously investigated social contagion and external factors. Professional proximity due to common patients does influence adoption decisions; geographical proximity is also influential, but its effect is more on rural than urban physicians. CONCLUSIONS: Flow of patients among different groups of physicians is a powerful factor in HIE adoption. Rather than merely following the market trend, physicians appear to be influenced by other physicians with whom they interact with and have common patients.

Commodities such as cloud resources (storage, computing, bandwidth) are often sold to clients on a pay-as-you-go basis. Thus, resource providers absorb all risk arising from end users' demand volatilities. We focus on the revenue risk management of commodities with highly volatile demand profiles using cloud computing as the application domain and bandwidth as the exemplar commodity. We extend the state of the art in risk hedging by introducing a new concept of dynamic forward contracts where a provider and a client flexibly interact through offers and responses over a set of time periods in a horizon. We develop an optimal pricing mechanism that takes into account the risk propensities of the provider and the client. The overall mechanism is modeled as a pair of nested dynamic programs denoting the offer-response interactions. The mechanism also incorporates two learning components: short-term learning on the client's demand and long-term learning on the client's risk propensity. We characterize two approaches for predicting the client's demand—a recursive demand prediction model and an aggregate demand prediction model. Detailed experimental studies of the proposed mechanism using real Web traffic data on the clients of Amazon Web Services have been carried out. The empirical results clearly demonstrate the superiority of the proposed mechanism over benchmark mechanisms such as the current industry practice of spot markets and static forward pricing mechanisms proposed in the literature in ex ante and ex post settings. The results also highlight key interaction effects among parameters controllable by a provider and the risk propensities of the market players, leading to valuable managerial implications for the practical adoption of the proposed mechanism.

Replication is one of the primary techniques used to improve the quality of distributed content service. It generally reduces user latencies and increases a site's availability. However, to our knowledge, there is no systematic framework that combines the structure of both content and service components of a Web application to design effective replica hosting architectures. Recent advances in interconnected and multiple content distribution network (CDN) architectures render this problem even more complex. In this study, we develop a systematic framework for designing and evaluating large-scale, component-based replication architectures for Web systems that are driven by both the quality and effectiveness of service provisioning on the service network. The proposed framework employs a combination of problem decomposition, configuration evaluation through controlled system simulations, and a neural-network-based feedback learning mechanism in the exploration of the design space. A case study demonstrates the viability of the framework. The framework can be an effective decision support tool for a system designer to systematically explore design options and select an appropriate design configuration that best meets the desired design objectives.

Integrated design of a large-scale distributed system is challenging due to multiple conflicting design objectives, a large number of stochastic parameters, and a combinatorially large design space. Current approaches using decomposition and analysis of isolated subproblems could lead to suboptimization. We develop a framework for integrated design that combines analytical modeling with embedded simulations for large-scale infrastructure planning. The framework uses problem decomposition and a guided simulation methodology, which uses concepts of extrinsic and intrinsic guidance. These are integrated in a comprehensive framework and used to interlink the subproblems and guide the design process towards achieving overall objectives concurrently. A case study demonstrates viability of the framework. The methodology is efficient and flexible, and can be adapted for other problems where resource sharing and integrated system design is critical.

A conceptual modeling grammar should be based on the theory of ontology and possess clear ontological semantics to represent problem domain knowledge in a precise and consistent manner. In this paper, we follow the notion of ontological expressiveness and conduct an ontological analysis of a newly-developed conceptual modeling grammar termed MibML (Multiagent-based Integrative Business Modeling Language). The grammar is developed to respond to the emerging needs for a special-purpose conceptual modeling grammar for the MIBIS (multi-agent-based integrative business information systems) universe. We assign ontological semantics to the MibML constructs and their relationship using the BWW (Bunge Wand-Weber) model. This article provides a starting point to further develop ontological principles and step-by-step guidelines to ensure the straightforward mapping from domain knowledge into MibML modeling constructs.

On-demand streaming from a remote server through best-effort Internet poses several challenges because of network losses and variable delays. The primary technique used to improve the quality of distributed content service is replication. In the context of the Internet, Web caching is the traditional mechanism that is used. In this article we develop a new staged delivery model for a distributed architecture in which video is streamed from remote servers to edge caches where the video is buffered and then streamed to the client through a last-mile connection. The model uses a novel revolving indexed cache buffer management mechanism at the edge cache and employs selective retransmissions of lost packets between the remote and edge cache for a best-effort recovery of the losses. The new Web cache buffer management scheme includes a dynamic adjustment of cache buffer parameters based on network conditions. In addition, performance of buffer management and retransmission policies at the edge cache is modeled and assessed using a probabilistic analysis of the streaming process as well as system simulations. The influence of different endogenous control parameters on the quality of stream received by the client is studied. Calibration curves on the QoS metrics for different network conditions have been obtained using simulations. Edge cache management can be done using these calibration curves. ISPs can make use of calibration curves to set the values of the endogenous control parameters for specific QoS in real-time streaming operations based on network conditions. A methodology to benchmark transmission characteristics using real-time traffic data is developed to enable effective decision making on edge cache buffer allocation and management strategies.

Abstract A majority of critical decisions requires collaborative efforts among analysts to build situation awareness. Teams of decision makers frequently have to react to incoming facts and developing events in a timely fashion such that the consequences of the decisions made largely have a positive impact on a developing situation. This problem is further exacerbated due to the multitude of agencies involved in the decision‐making process. Thus, the decision‐making processes faced by the intelligence agencies are characterized by group deliberations that are highly ill structured and yield limited analytical tractability. In this context, a collaborative approach to providing cognitive support to decision makers using a connectionist modeling approach is proposed. The connectionist modeling of such decision scenarios offers several unique and significant advantages in developing systems to support collaborative discussions. Several inference rules for augmenting the argument network and to capture implicit notions in arguments are proposed. We further explore the effects of incorporating notions of information source reliability within arguments and the effects thereof.

The focus of this work is on join optimization in relational database systems. The importance of join optimization is critically underscored by the high cost of relational joins and their frequent needs in traditional as well as emerging database applications. We demonstrate that the sequence in which the pages of the relations are accessed to process a join is a critical determinant of the join execution cost and an optimization of this sequence can lead to a significant improvement in performance over the traditional approaches. We initially develop three network structures to represent a join on two relations: page connectivity graph, cascade and block tree cascade. A page connectivity graph is a bipartite representation of the set of connected pages in the two relations according to the join predicate. To reveal the structural properties of the join, the nodes of the bipartite graph are ordered into a set of levels, and the resulting isomorphic structure is termed a cascade. From the cascade, a tree structure termed a block tree cascade is derived by selectively grouping nodes at each level of the cascade into blocks. We formulate the join as a tree traversal process, and accordingly develop efficient tree traversal algorithms. We develop a compact data structure to store the resulting access path, and provide a comprehensive analysis of the algorithms with detailed assessments of their performance. The performance evaluation demonstrates that the proposed approach can result in significant cost savings over the current join processing methods, for low to modest values of the join selectivity factor.

Join processing in relational database systems continues to be a difficult and challenging problem. In this research, we propose a criss-cross hash join strategy that draws from both hashing and indexing techniques, inheriting the advantages of each. To facilitate the criss-cross hash join, a simple data structure, termed page map, is introduced. The page maps aid in reducing the hashing effort incurred in the current hash based join methods. Furthermore, the page maps implicitly capture and exploit the possible inherent order among tuples in the relations, however partial it may be, to achieve superior performance. As the proposed methodology relies on the hashing scheme, the page maps are simpler, more compact, and easier to maintain than the traditional data structures associated with index based join methods. We develop the ideas intuitively first, followed by a formal development of the concepts and the algorithms. A detailed probabilistic analysis of the algorithms is presented and their performance is assessed through extensive empirical investigations. The empirical analysis suggests significant performance improvements over the current state-of-the-art hybrid hash method, especially in the presence of possible inherent order.

One of the challenges faced by an enterprise with employees in multiple locations is to design a high performance, secure and interoperable distributed computing system (DCS) to interconnect all locations and operations. The problem is computationally hard; hence, subcomponents of it have been studied in detail. These tend to be complex, predominantly theoretical, and somewhat limited from a practical standpoint in terms of providing an integrated solution. Researchers have indicated the need to study the integrated DCS design problem. We merge the complex components and demonstrate that this intrinsically hard problem can be modeled using a component-wise iterative approach. We present a methodology for resource planning and integration that is simple and practical, and can be applied to real-life problems. The methodology includes performance, cost, security and interoperability issues as DCS design objectives. It distributes data and application systems across multiple locations, and aids in security and interoperability configuration, such that the overall design objectives are satisfied. This is one of the first attempts at combining the various components of DCS design and applying it to a real-life problem. The design framework has been successfully used to design a distributed training system for a large, geographically dispersed organization.

In this paper, we develop a unified framework for approximating open and closed queueing networks under any general blocking protocol by extending and generalizing the approximation algorithm for open tandem queues under minimal blocking presented in Di Mascolo et al. (1996). The proposed framework is based on decomposition. We develop decomposition structures and analysis algorithms for any general blocking system using the framework. The proposed algorithms have been extensively tested using simulations as a benchmarking device. The results show that the proposed framework yields robust, reliable, and accurate estimates of system characteristics, such as throughput and Work-In-Process inventory in a wide range of system configurations. The computational load is minimal. The unified framework presents a highly useful set of tools of analysis for queueing-system designers to use in evaluating the performance under numerous design alternatives. Directions for future research are presented, with a focus on critical application areas such as packet-switching-network design and cellular manufacturing.

The objective of access path optimization in relational joins is to minimize the number of pages accessed in performing a join, since the cost of data retrieval from disk accounts for most of the join processing costs. We develop an efficient approach to access path optimization by modeling an access path as a tree traversal process. Two join algorithms are developed using this model. The access path is stored as a compact index relation, which requires absolutely minimal storage. The join is easily materialized by reading the index relation sequentially, and performing the operations coded in its tuples in their serial order. The proposed approach is shown to result in globally minimal page accesses in certain classes of joins. Using a detailed probabilistic analysis and computational study, the proposed approach is shown to be much superior to the state-of-the-art join optimization methods for a general class of joins. INFORMS Journal on Computing, ISSN 1091-9856, was published as ORSA Journal on Computing from 1989 to 1995 under ISSN 0899-1499.

In this research, we address the query clustering problem which involves determining globally optimal execution strategies for a set of queries. The need to process a set of queries together often arises in deductive database systems, scientific database systems, large bibliographic retrieval systems and several other database applications. We address the optimization problem from the perspective of overlaps in data requirements, and model the batched operations using a set-partitioning approach. In this model, we first consider the case of m queries each involving a two-way join operation. We develop a recursive methodology to determine all the processing strategies in this case. Next, we establish certain dominance properties among the strategies, and develop exact as well as heuristic algorithms for selecting an appropriate strategy. We extend this analysis to a clustering approach, and outline a framework for optimizing multiway joins. The results show that the proposed approach is viable and efficient, and can easily be incorporated into the query processing component of most database systems.

An efficient interactive solution framework for bicriteria integer programming is developed. The proposed methodology follows the implicit utility maximization approach. The decision maker's underlying utility function is assumed to be pseudoconcave and nondecreasing, and the problem is solved using an interactive branch-and-bound methodology. Several new concepts on bicriteria integer programming that offer great efficiency in the solution process are developed. The framework has been tested extensively, and results with problems having up to 80 variables and 40 constraints are presented. The results show that the methodology is an effective approach to solving practical bicriteria problems.>