Enforcing Service-Level Constraints in Supply Chains With Assembly Operations

We study a feedforward supply network that involves assembly operations. We compute optimal stock levels which minimize inventory costs and maintain stockout probabilities below given desirable levels (service-level constraints). To that end, we develop large deviations approximations for inventory costs and service level constraints and formulate the stock level selection problem as a nonlinear programming problem which can be solved using standard techniques. This results in significant computational savings when compared to exhaustive search using simulation. Our distributional assumptions are general enough to include temporal dependencies in the demand and production processes. We leverage the solution of the inventory control problem in the design of supply contracts under explicit service-level constraints     

Learning from Network Device Statistics

We estimate end-to-end service metrics from network device statistics. Our approach is based upon statistical, supervised learning, whereby the mapping from device-level to service-level metrics is learned from observations, i.e., through monitoring the system. The approach enables end-to-end performance prediction without requiring an explicit model of the system, which is different from traditional engineering techniques that use stochastic modeling and simulation. The fact that end-to-end service metrics can be estimated from local network statistics with good accuracy in the scenarios we consider suggests that service-level properties are “encoded” in network-level statistics. We show that the set of network statistics needed for estimation can be reduced to a set of measurements along the network path between client and service backend, with little loss in estimation accuracy. The reported work is largely experimental and its results have been obtained through testbed measurements from a video streaming service and a KV store over an OpenFlow network .     

基于模糊层次分析的网络服务级安全态势评价方法

针对缺乏网络服务级安全态势评价方法的现状,利用事件注入技术,充分考虑影响服务可用性与性能的重要因素,确立了反映网络服务级安全态势的三级指标体系,并提出一种基于模糊层次分析法(FAHP)的网络服务级安全态势评价方法,通过求出各指标对上一级指标的权重,结合多层次综合评价方法得到模糊最大隶属度向量,从而对网络服务级安全态势进行了量化分析。实验结果表明该方法能有效屏蔽具体服务及入侵行为细节,实现对网络服务级安全态势的定性描述和定量分析。     

Deploying Diffserv at the network edge for tight SLAs, Part 2

In the second of a two-part series, the authors review industry best practices for designing, validating, deploying, and operating IP-based services at the network edge with tight service-level agreements (SLAs). Specifically, they present a case study that shows how Diffserv can be deployed to achieve these SLAs.     

Modeling and Verification of Reliable Messaging by Graph Transformation Systems

Due to the increasing need of highly dependable services in Service-Oriented Architectures (SOA), service-level agreements include more and more frequently such non-functional aspects as security, safety, availability, reliability, etc. Whenever a service can no longer be provided with the required QoS, the service requester needs to switch dynamically to a new service having adequate service parameters after exchanging a sequence of messages. In the current paper, we first extend the core SOA metamodel with parameters required for reliable messaging in services. Then we model reconfigurations for reliable message delivery by graph transformation rules. Finally, we carry out a formal verification of the proposed rule set by combining analysis tools for graph transformation and labeled transition systems.     

Web服务器集群系统QoWS研究

请求负载的增加常常导致Web服务器系统性能降低,用户期望的服务质量得不到保证,这是服务级Web系统所面临和必须解决的问题。文章提出了一种Web服务器集群环境下的负载分配策略,通过对用户请求分类、将不同类别的请求进行响应性能隔离、优先为高级别请求提供服务以及请求许可控制等手段,对不同类别的Web请求提供不同质量的服务,保证了服务级用户的服务质量。同时采用最迟分配原则,改善系统的负载均衡能力,缩短系统平均响应时间。最后通过仿真实验,验证了该策略的正确性和有效性。     

Deploying Diffserv at the network edge for tight SLAs, part I

In the first of a two-part series, we review industry best practices for designing, validating, deploying, and operating IP-based services at the network edge with tight service-level agreements (SLAs). We describe the important SLA metrics for IP service performance and discuss why Diffserv is the preferred technology to achieve these SLAs.     

Approximation algorithms for minimizing the total weighted number of late jobs with late deliveries in two-level supply chains

We study a supply chain scheduling problem in which n jobs have to be scheduled on a single machine and delivered to m customers in batches. Each job has a due date, a processing time and a lateness penalty (weight). To save batch-delivery costs, several jobs for the same customer can be delivered together in a batch, including late jobs. The completion time of each job in the same batch coincides with the batch completion time. A batch setup time has to be added before processing the first job in each batch. The objective is to find a schedule which minimizes the sum of the weighted number of late jobs and the delivery costs. We present a pseudo-polynomial algorithm for a restricted case, where late jobs are delivered separately, and show that it becomes polynomial for the special cases when jobs have equal weights and equal delivery costs or equal processing times and equal setup times. We convert the algorithm into an FPTAS and prove that the solution produced by it is near-optimal for the original general problem by performing a parametric analysis of its performance ratio.     

Automated maintenance of service compositions with SLA violation detection and dynamic binding

Web service compositions need to adapt to changes in their constituent web services, in order to maintain functionality and performance. Therefore, service compositions must be able to detect web service failure and performance degradation resulting in the violation of service-level agreements. Automated diagnosis and repair are equally important. However, existing standards and languages for service compositions, such as BPEL, lack constructs for web service monitoring and runtime adaptability, which are pre-requisites for diagnosis and repair. We present a solution for transparent runtime monitoring, as well as automated performance degradation detection, diagnosis, and repair for service compositions expressed as BPEL processes. Our solution uses lightweight monitoring techniques, supports customizable diagnosis and repair strategies, and is compatible with any standards-compliant BPEL engine.     

Service-level agreement–aware scheduling and load balancing of tasks in cloud

Cloud computing is an innovative computing paradigm designed to provide a flexible and low-cost way to deliver information technology services on demand over the Internet. Proper scheduling and load balancing of the resources are required for the efficient operations in the distributed cloud environment. Since cloud computing is growing rapidly and customers are demanding better performance and more services, scheduling and load balancing of the cloud resources have become very interesting and important area of research. As more and more consumers assign their tasks to cloud, service-level agreements (SLAs) between consumers and providers are emerging as an important aspect. The proposed prediction model is based on the past usage pattern and aims to provide optimal resource management without the violations of the agreed service-level conditions in cloud data centers. It considers SLA in both the initial scheduling stage and in the load balancing stage, and it looks into different objectives to achieve the minimum makespan, the minimum degree of imbalance, and the minimum number of SLA violations. The experimental results show the effectiveness of the proposed system compared with other state-of-the-art algorithms.     

Optimal Threshold Levels in Stochastic Fluid Models via Simulation-based Optimization

A number of important problems in production and inventory control involve optimization of multiple threshold levels or hedging points. We address the problem of finding such levels in a stochastic system whose dynamics can be modelled using generalized semi-Markov processes (GSMP). The GSMP framework enables us to compute several performance measures and their sensitivities from a single simulation run for a general system with several states and fairly general state transitions. We then use a simulation-based optimization method, sample-path optimization, for finding optimal hedging points. We report numerical results for systems with more than twenty hedging points and service-level type probabilistic constraints. In these numerical studies, our method performed quite well on problems which are considered very difficult by current standards. Some applications falling into this framework include designing manufacturing flow controllers, using capacity options and subcontracting strategies, and coordinating production and marketing activities under demand uncertainty.     

User utility discovery for priority-based network resource pricing

While equilibrium analysis has been commonly used for network pricing under the assumption that user utility functions are precisely known, many researchers have criticized the validity of the assumption. In this paper, we propose a solution for bridging the gap between the existing theoretical work on optimal pricing and the unavailability of precise user utility information in real networks. In the proposed method, the service provider obtains increasingly more accurate estimates of user utility functions by iteratively changing the prices of service levels and observing the users’ service-level choices under various prices. Our study’s contribution is twofold. First, we have developed a general principle for estimating user utility functions. Especially, we present the utility estimation for dynamic user population. Second, we have developed a method for setting prices that can optimize the extraction of information about user utility functions. The extensive simulation results demonstrate the effectiveness of our method.     

WSDL and UDDI extensions for version support in web services

Versioning is an important aspect of web service development, which has not been adequately addressed so far. In this article, we propose extensions to WSDL and UDDI to support versioning of web service interfaces at development-time and run-time. We address service-level and operation-level versioning, service endpoint mapping, and version sequencing. We also propose annotation extensions for developing versioned web services in Java. We have tested the proposed solution for versioning in two real-world environments and identified considerable improvements in service development and maintenance efficiency, improved service reuse, and simplified governance.     

Group rekeying with limited unicast recovery

In secure group communications, a key server can deliver a “group-oriented” rekey message [C.K. Wong, M.G. Gouda, S.S. Lam, Secure group communications using key graphs, in: Proceedings of ACM SIGCOMM ’98, September 1998, pp. 68–79] to a large number of users efficiently using multicast. For reliable delivery, Keystone [C.K. Wong, S.S. Lam, Keystone: a group key management system, in: Proceedings of International Conference on Telecommunications, Acapulco, Mexico, May 2000] proposed the use of forward error correction (FEC) in an initial multicast, followed by the use of unicast delivery for users that cannot recover their new keys from the multicast. In this paper, we investigate how to limit unicast recovery to a small fraction r of the user population. By specifying a very small r, almost all users in the group will receive their new keys within a single multicast round.We present analytic models for deriving r as a function of the amount of FEC redundant information (denoted by h) and the rekeying interval duration (denoted by T) for both Bernoulli and two-state Markov Chain loss models. From our analyses, we conclude that r decreases roughly at an exponential rate as h increases. We then present a protocol designed to adaptively adjust (h,T) to achieve a specified r. In particular, our protocol chooses from among all feasible (h,T) pairs one with h and T values close to their feasible minima. Our protocol also adapts to an increase in network traffic. Simulation results using ns-2 show that with network congestion our adaptive FEC protocol can still achieve a specified r by adjusting values of h and T.     

A tale of two pricing systems for services

Due to advances in technology and the rapid growth of online service offerings, various innovative web-based service models and delivery methods have appeared—including several free services. It is not always clear whether and how these emerging mechanisms for online service delivery will result in profitable businesses. In this paper, with an eye towards beginning to understand the issues involved, we present an analytical model of rational customer choice between available service plans. In particular, our model predicts how a monopoly service provider should devise its plans, if it understands such customer behavior. We then describe how this model would need to be extended in order to reflect increasingly inexpensive and even free service offerings.     

Performance and availability of Internet data centers

We use an Internet data center (IDC) as a motivating example to discuss how performance and availability are interrelated. IDCs provide the means for geographically distributed Internet users to tap into other computers and applications. IDC users pay for the services they obtain, so they want them to be delivered according to established service-level agreements (SLAs). These agreements indicate service performance levels and availability. An IDC must provision enough capacity and redundant resources to ensure that it can meet its performance and availability SLAs.     

A risk-averse approach to simulation optimization with multiple responses

This article considers risk-averse simulation optimization problems, where the risk measure is the well-known Average Value-at-Risk (also known as Conditional Value-at-Risk). Furthermore, this article combines Taguchi’s robustness with Response Surface Methodology (RSM), which results in a novel, robust RSM to solve such risk-averse problems. In case the risk-averse problem is convex, the conic quadratic reformulation of the problem is provided, which can be solved very efficiently. The proposed robust RSM is applied to both an inventory optimization problem with a service-level constraint and a call-center problem; the results obtained for the risk-averse problem and its benchmark problem, where the risk measure is the variance, are intuitively sound and interesting.     

Coupled Path Planning,Region Optimization,and Applications in Intensity-modulated Radiation Therapy

In this paper, we consider an optimization problem in discrete geometry, called coupled path planning (CPP). Given a finite rectangular grid and a non-negative function f defined on the horizontal axis of the grid, we seek two non-crossing monotone paths in the grid, such that the vertical difference between the two paths approximates f in the best possible way. This problem arises in intensity-modulated radiation therapy (IMRT), where f represents an ideal radiation dose distribution and the two coupled paths represent the motion trajectories (or control sequence) of two opposing metal leaves of a delivery device for controlling the area exposed to the radiation source. By finding an optimal control sequence, the CPP problem aims to deliver precisely a prescribed radiation dose, while minimizing the side-effects on the surrounding normal tissue. We present efficient algorithms for different versions of the CPP problems. Our results are based on several new ideas and geometric observations, and substantially improve the solutions based on standard techniques. Implementation results show that our CPP algorithms run fast and produce better quality clinical treatment plans than the previous methods.     

Network-aware partial caching for Internet streaming media

The delivery of multimedia over the Internet is affected by adverse network conditions such as high packet loss rate and long delay. This paper aims at mitigating such effects by leveraging client-side caching proxies. We present a novel cache architecture and associated cache management algorithms that turn edge caches into accelerators of streaming media delivery. This architecture allows partial caching of media objects and joint delivery from caches and origin servers. Most importantly, the caching algorithms are both network-aware and stream-aware; they take into account the popularity of streaming media objects, their bit rate requirements, and the available bandwidth between clients and servers. Using Internet bandwidth models derived from proxy cache logs and measured over real Internet paths, we have conducted extensive simulations to evaluate the performance of various cache management algorithms. Our experiments demonstrate that network-aware caching algorithms can significantly reduce startup delay and improve stream quality. Our experiments also show that partial caching is particularly effective when bandwidth variability is not very high.Shudong Jin: Corespondence to This research was supported in part by NSF (awards ANI-9986397, ANI-0095988, ANI-0205294 and EJA-0202067) and by IBM. Part of this work was done while the first author was at IBM Research in 2001.