Guaranteed Service Level Research Articles

Analytical architecture-based performability evaluation of real-time software systems

Real-time systems are usually employed in dynamic and harsh environments. Real-time software, as one important part of such systems, potentially suffers from two problems: unpredictability in the timing behavior which affects the software performance, and logical faults which affect the software reliability. The former problem is mitigated by improving the software algorithm, architecture, and code. The latter problem is also relieved via software redundancy methods, even though these methods may adversely affect the software performance and architectural complexity. Despite these problems, it is expected to have a guaranteed service level in real-time systems, which the service is defined as the successful completion of the software mission within its deadline. In this paper, we propose two architecture-based analytical methods for simultaneous performance and reliability (performability) evaluation of real-time component-based software: one is accurate and the other is approximate. The accurate method is sound and precise but more complex in the computations, while the approximate method is easy-to-follow with reasonable amounts of computations. Examples of different configurations have been presented to show how well the latter method approximates the former one. Some performability sensitivity analyses with respect to the software component properties have also been done for better depiction of the importance of employing the proposed analytical methods in finding and eliminating the software performability bottlenecks.

Dual time-scale distributed capacity allocation and load redirect algorithms for cloud systems

Resource management remains one of the main issues of cloud computing providers because system resources have to be continuously allocated to handle workload fluctuations while guaranteeing Service Level Agreements (SLA) to the end users. In this paper, we propose novel capacity allocation algorithms able to coordinate multiple distributed resource controllers operating in geographically distributed cloud sites. Capacity allocation solutions are integrated with a load redirection mechanism which, when necessary, distributes incoming requests among different sites. The overall goal is to minimize the costs of allocated resources in terms of virtual machines, while guaranteeing SLA constraints expressed as a threshold on the average response time. We propose a distributed solution which integrates workload prediction and distributed non-linear optimization techniques. Experiments show how the proposed solutions improve other heuristics proposed in literature without penalizing SLAs, and our results are close to the global optimum which can be obtained by an oracle with a perfect knowledge about the future offered load.

Competition and Contracting in Service Industries

Two very different contractual structures are commonly observed in service industries with congestion effects: service level guarantees (SLGs) and best effort (BE) service. We analyze the impact of these contractual agreements on market outcomes in oligopolistic industries. First, we consider a model where firms compete by setting prices and SLGs simultaneously. The SLG is a contractual obligation on the part of the service provider: regardless of how many customers subscribe, the firm is responsible for investing so that the congestion experienced by all subscribers is equal to the SLG. We then consider the BE contractual model where firms compete by setting prices and investment levels simultaneously. With BE contractual agreements, firms provide the best possible service given their infrastructure, but without an explicit guarantee. Using the Nash equilibria (NE) of the games played by firms, we compare these competitive models in terms of the resulting prices, service levels, firms' profits, and consumers' surplus. We first show that the SLG game can be reduced to a standard pricing game, greatly simplifying the analysis of this otherwise complex competitive scenario. We then compare the SLG game with the BE game; equilibria for the BE is characterized in a previous paper. Using these results we show that in the case of constant returns to investment, while the NE price for the SLG game is perfectly competitive, firms obtain positive markups in the unique NE for the BE game. We also study the firms' choice of the strategy space, i.e., whether to offer SLG or BE contracts to the consumer, and find that competition is intensified if even one firm chooses to offer SLG contracts. Our results contribute to the basic understanding of competition and contracting in service industries and yield insight into business and policy considerations.

Robust Airline Scheduling Under Block-Time Uncertainty

Airline schedule development continues to remain one of the most challenging planning activities for any airline. An airline schedule comprises a list of flights and specifies the origin, destination, scheduled departure, and arrival time of each flight in the airline's network. A critical component of the schedule development activity is the choice of flight block-times, which depend on several factors. Many airlines decide schedule block-times based on fixed percentiles of block-time distributions built from historical data, however, such techniques have not resulted in significantly improved on-time performance (OTP) of the schedule during operations. Thus, from a passenger's perspective, the service-level guarantee of an airline's network continues to be low. We first define two service-level metrics for an airline schedule. The first one is similar to the OTP measure of the U.S. Department of Transportation and we define it as the flight service level. The second metric, called the network service level, is geared toward completion of passenger itineraries. We then develop a stochastic integer programming formulation that optimally perturbs a given schedule to maximize expected profit, while ensuring the two service levels. We also develop a variant of this model that maximizes service levels, while achieving desired network profitability. To solve these models, we develop an efficient algorithm that guarantees optimality. Through extensive computational experiments, using real-world data, we demonstrate that our models and algorithms are efficient and achieve the desired trade-off between service level and profitability.

Competition and contracting in service industries

In service industries with congestion effects, two very different contractual structures are commonly observed, depending on whether or not firms choose to offer a guaranteed service level. We analyze the impact of these choices on market outcomes in oligopolistic industries. Our results highlight how different contractual agreements change the intensity of price competition in service industries. Broadly speaking, we show that competition is intensified when firms choose to offer service level guarantees.

Sustainable Development of Maritime Operations in Ports

ABSTRACTGlobalisation has heralded burgeoning ship movements and maritime operations in ports alongside increased international concerns regarding potential environmental impacts. In particular, smaller ports require accessible tools to manage them. A framework to facilitate environmental management applies business process principles to identify relevant inputs, processes and outputs. A case study of Falmouth Harbour Commissioners compares functional units and flows that define input–output processes for anchoring and bunkering operations. Strategic‐level processes affect present and future operations while tactical service processes guarantee service level and quality through their integrity. Operational processes occur at the output level. An accessible generic framework supports planning of more sustainable maritime operations, facilitates mitigation of potential risks and encourages authorities to engage with sustainability agendas and manage development proposals proactively. Ongoing interlocution with business strategists will refocus port managers on educational and commercial missions and increase stakeholder engagement. Simplification and optimisation phases of business process re‐engineering remain untapped by business strategists. Copyright © 2011 John Wiley & Sons, Ltd and ERP Environment.

Inter-Cloud Computing

This catchword article presents the concept of “Inter-Cloud Computing” which introduces an additional management layer on top of conventional Cloud Computing Systems. Its goal is to reach a higher level of sustainability by autonomously shifting resources among the participating Cloud systems when unexpected load levels occur or disasters strike. Cloud systems are widely seen as a promising paradigm for an IT infrastructure that is capable of creating an added value for business, society, and administration. If Cloud systems are to be applied to critical areas, such as enterprises’ core businesses, governmental systems, medical applications, and other social infrastructure services, it is crucial to guarantee end-to-end service quality that covers the involved external networks, and to meet a set of requirements regarding reliability (including security), compliance, governance, and power efficiency. However, if services are provided by a single Cloud system – which is the usual case today – unexpected levels of overload traffic from the Internet or natural disasters may affect the system’s reliable functioning. The ability of a single Cloud to request reserved resources is usually fixed by limited bounds. Therefore, unexpected loads and failures can easily overburden a single Cloud system and lead to unreliable and interrupted services. In order to enable a Cloud system to continue the delivery of guaranteed service levels even in such cases, it is indispensable to introduce a super-ordinate layer of Cloud organization which manages multiple Cloud systems so that they complement each other. Thus single Cloud systems, which are interconnected via broadband networks, can mutually request required resources from their peers, or provide available capacity to them, respectively. This flavor of Cloud Computing is called Inter-Cloud Computing.

Inter-Cloud-Computing

This catchword article presents the concept of “Inter-Cloud Computing” which introduces an additional management layer on top of conventional Cloud Computing Systems. Its goal is to reach a higher level of sustainability by autonomously shifting resources among the participating Cloud systems when unexpected load levels occur or disasters strike. Cloud systems are widely seen as a promising paradigm for an IT infrastructure that is capable of creating an added value for business, society, and administration. If Cloud systems are to be applied to critical areas, such as enterprises’ core businesses, governmental systems, medical applications, and other social infrastructure services, it is crucial to guarantee end-to-end service quality that covers the involved external networks, and to meet a set of requirements regarding reliability (including security), compliance, governance, and power efficiency. However, if services are provided by a single Cloud system – which is the usual case today – unexpected levels of overload traffic from the Internet or natural disasters may affect the system’s reliable functioning. The ability of a single Cloud to request reserved resources is usually fixed by limited bounds. Therefore, unexpected loads and failures can easily overburden a single Cloud system and lead to unreliable and interrupted services. In order to enable a Cloud system to continue the delivery of guaranteed service levels even in such cases, it is indispensable to introduce a super-ordinate layer of Cloud organization which manages multiple Cloud systems so that they complement each other. Thus single Cloud systems, which are interconnected via broadband networks, can mutually request required resources from their peers, or provide available capacity to them, respectively. This flavor of Cloud Computing is called Inter-Cloud Computing.

Loosely coupled coordinated management in virtualized data centers

Management is an important challenge for future enterprises. Previous work has addressed platform management (e.g., power and thermal management) separately from virtualization management (e.g., virtual machine (VM) provisioning, application performance). Coordinating the actions taken by these different management layers is important and beneficial, for reasons of performance, stability, and efficiency. Such coordination, in addition to working well with existing multi-vendor solutions, also needs to be extensible to support future management solutions potentially operating on different sensors and actuators. In response to these requirements, this paper proposes vManage, a solution to loosely couple platform and virtualization management and facilitate coordination between them in data centers. Our solution is comprised of registry and proxy mechanisms that provide unified monitoring and actuation across platform and virtualization domains, and coordinators that provide policy execution for better VM placement and runtime management, including a formal approach to ensure system stability from inefficient management actions. The solution is instantiated in a Xen environment through a platform-aware virtualization manager at a cluster management node, and a virtualization-aware platform manager on each server. Experimental evaluations using enterprise benchmarks show that compared to traditional solutions, vManage can achieve additional power savings (10% lower power) with significantly improved service-level guarantees (71% less violations) and stability (54% fewer VM migrations), at low overhead.

Scalable voice over Internet protocol service-level agreement guarantees in converged transmission control protocol/Internet protocol networks

Increasing complexity of the recent introduced quality of service (QoS) provisioning tools in packet-switched networks becomes a major obstacle for Internet service providers to practically deploy and manage these mechanisms to guarantee a range of transparent real-time service-level agreements. The authors revisit the existing Internet protocol (IP) QoS approaches and present a cost-effective solution to provide voice over Internet protocol (VoIP) service guarantees in converged IP networks by combining open shortest-path first traffic engineering with new insights for the configuration of standard active queue management random early detection mechanism. Extensive simulation studies are carried out, and the service quality is explicitly quantified using international telecommunication union E-model for a range of scenarios representing different types of network uncertainties. The direct representation of VoIP service quality well demonstrate that such a simple, coordinated approach, in keeping with the Internet paradigm, can achieve increased load for a given quality level and greater resilience under degraded network conditions.

Dynamic wavelength and bandwidth allocation using adaptive linear prediction in WDM/TDM ethernet passive optical networks

Hybrid wavelength-division-multiplexing (WDM)/time-division-multiplexing (TDM) ethernet passive optical networks (EPONs) can achieve low per-subscriber cost and scalability to increase the number of subscribers. This paper discusses dynamic wavelength and bandwidth allocation (DWBA) algorithm in hybrid WDM/TDM EPONs. Based on the correlation structure of the variable bit rate (VBR) video traffic, we propose a quality-of-service (QoS) supported DWBA using adaptive linear traffic prediction. Wavelength and timeslot are allocated dynamically by optical line terminal (OLT) to all optical network units (ONUs) based on the bandwidth requests and the guaranteed service level agreements (SLA) of all ONUs. Mean square error of the predicted average arriving rate of compound video traffic during waiting period is minimized through Wiener-Hopf equation. Simulation results show that the DWBA-adaptive-linear-prediction (DWBA-ALP) algorithm can significantly improve the QoS performances in terms of low delay and high bandwidth utilization.

Price competition with service level guarantee in web services

Web services have become quite popular over the last few years as they allow easier development and integration of business applications. Unlike traditional software systems, web services are self-contained modular software components that are delivered over a network (such as the Internet) and executed on a remote system hosting the requested services. However, the network and processing overhead associated with web services have also presented a significant challenge to its performance. As a result, a web service provider often announces a service-level agreement when launching a service. The service-level agreement provides a guarantee to the consumers that they can get the service they pay for at an assured level of quality. In this paper, we study the competition between two such providers offering functionally the same web services. Each provider needs to decide a service level (standard or premium) she would offer and a corresponding price for the selected service level to meet the QoS guarantee (in terms of an average response time of the service). We first analyze the case where the providers choose service levels and prices simultaneously, and then extend it to a sequential-move situation. Finally, we examine strategic choices of providers when the processing capacity is endogenized into the model.

Measurement of packet loss probability by optimal design of packet probing experiments

Packet-level measurement is now critical to many aspects of broadband networking, for example for guaranteeing service level agreements, facilitating measurement-based admission control algorithms and performing network tomography. Because it is often impossible to measure all the data passing across a network, the most widely used method of measurement works by injecting probe packets. The probes provide samples of the packet loss and delay, and from these samples the loss and delay performance of the traffic as a whole can be deduced. However, measuring performance like this is prone to errors. Recent work has shown that some of these errors are minimised by using a gamma renewal process as the optimal pattern for the time instants at which to inject probes. This leaves the best rate at which to inject probes as the key unsolved problem, and this is addressed here by using the statistical principles of the design of experiments. The experimental design approach allows one to treat packet-level measurements as numerical experiments that can be designed optimally. Modelling the overflow of buffers as a 2-state Markov chain, the system's likelihood function is deduced, and from this a technique (using the Fisher information matrix) to determine the upper-bound on the optimal rate of probing is developed. A generalisation of this method accounts for the effect of the probed observations interfering with the experiment. The numerical results focus on VoIP traffic, allowing one to show how this methodology would be used in practice. One application of this is in measurement-based admission control algorithms, where the technique can be used to provide an upper-bound on the rate at which probes should be injected to monitor the loss performance of the target network, prior to making an admit/do not admit decision.

The stochastic [formula omitted]-hub center problem with service-level constraints

Small package delivery companies offer services where packages are guaranteed to be delivered within a given time-frame. With variability in travel time, the configuration on the hub-and-spoke delivery network is vital in ensuring a high probability of meeting the service-level guarantee. We present the stochastic p -hub center problem with chance constraints, which we use to model the service-level guarantees. We discuss analytical results, propose solution heuristics, and present the results from computational experiments.

Policy Adjuster-driven Grid Workflow Management for Collaborative Heart Disease Identification System

Abstract: This paper proposes a policy adjuster-driven Grid workflow management system for collaborative healthcare platform, which supports collaborative heart disease diagnosis applications. To select policies according to service level agreement of users and dynamic resource status, we devised a policy adjuster to handle workflow management polices and resource management policies using policy decision scheme. We implemented this new architecture with workflow management functions based on policy quorum based resource management system for providing poincare geometry-characterized ECG analysis and virtual heart simulation service. To evaluate our proposed system, we executed a heart disease identification application in our system and compared the performance to that of the general workflow system and PQRM system under different types of SLA. Keywords: Grid Worfklow, Collaborative Healthcare Platform, SLA, Policy Adjuster 1. Introduction There have been many attempts to create healthcare system or platform by combining the existing medical service with IT such as sensor and communication technology. In other words, the general health clinical data and information are exchanged electronically, and the patient, doctor, hospital and laboratories can be connected through communication network to perform medical practices. Such healthcare system is enabled and supported by IT, yet the core technology will be high-speed information exchange technology and high performance computing technology in distributed environment. Grid technology can meet these requirements by bringing heterogeneous resources together and allocating them efficiently to applications. Medical applications are no longer restricted to one medical institute. Some collaborated applications like heart disease simulator, an instance of workflow jobs, require several resources work together. So the healthcare platform requires collaborative job scheduling function based on resource status which is the core part of Grid resource management system [2][3]. However, the complexity of resource management for QoS guarantee in Grid significantly increases as the grid computing grows. One of the promising approaches trying to solve this problem is policy based resource management system [2] that is suitable to the complex management environments. To manage resource in Grid, it is required to have a sort of grid resource broker and scheduler that manage resources and jobs in the virtual Grid environment. Authors in [3] suggested policy-based resource management system (PQRM) architecture by consideration of Grid Service Level Agreement (SLA) and abstract architecture of grid resource management. In medical healthcare environment, beside most of single-program applications, there exist many applications which require co-process of many programs following a strict processing order. Since these applications are executed on Grid, they are called Grid workflows. Current Grid resource management systems do not consider complexity of collaborative medical applications and service level guarantee at workflow level. In this paper, we propose Grid workflow-driven healthcare platform for collaborative applications. In the sense of optimum management, PQRM [5] provides optimum resource quorum for applications, so we develop workflow-Integrated PQRM as one instance for this workflow-driven healthcare platform. PQRM’s available resource quorum is optimum resource set in resource management point of view and our workflow-driven system can provide optimum resource set in workflow management point of view. Two management philosophies require a policy decision scheme to adjust based on user’s SLA. We evaluate the proposed workflow-driven healthcare platform by measuring the completion time of collaborative heart disease simulator under different types of user SLA. DOI : 10.3745/JIPS.2008.4.3.103

A particle swarm optimization based algorithm to the Internet subscription problem

It is a business practice that Internet service providers (ISPs) employ the differential service concept to satisfy various requirements of different Internet customers. Under this concept, the bandwidth resources are usually divided into several Internet commodity classes for sales. For each commodity class, there is a different sales price and a unique bandwidth restriction. Theoretically, subscribers can enjoy their services up to the bandwidth limits. However, in reality, many subscribers may encounter a situation in which their on-line requests are denied or their connection speeds are far below their contracted speed limits. One of the reasons is that ISPs do not set sales limits for their Internet commodities and accept too many subscribers as their Internet subscribers. To overcome this problem, this paper deals with an ISP’s sales limit problem over a finite time planning period. The purpose of this paper is to maximize the total expected revenues by determining the sales limits for each subscription type and bandwidth allocation for each commodity class over any subscription interval under service level guarantee. In this paper, a constrained nonlinear integer model is developed and an algorithm based on a particle swarm optimization (PSO) is proposed to acquire admissible solutions. Several examples are provided for showing the efficiency of the proposed method.

Competition in Prices and Service Level Guarantees

In this paper we study the implications of service level guarantees (SLGs) in a model of oligopoly competition where providers compete to deliver a service to congestion-sensitive consumers. The SLG is a contractual obligation on the part of the service provider: regardless of how many customers subscribe, the firm is responsible for investing in infrastructure, capacity, or service quality so that the congestion experienced by all subscribers is equal to the SLG. First, we analyze a game where firms compete by setting prices and SLGs simultaneously. We establish that this game can be reduced to standard oligopoly models of price competition, greatly simplifying the analysis of this otherwise complex competitive scenario. Notably, we find that when costs in the original game are convex, the resulting equivalent pricing game also has convex costs. Further, for a broad class of models exhibiting constant returns to investment, the resulting pricing game is equivalent to a standard price game with constant marginal costs; many loss systems, such as those modeled by the Erlang loss formula, exhibit constant returns to investment. We then consider another commonly used contractual agreement between firms and customers: firms first set prices and investment levels simultaneously, and then consumers choose where to subscribe. In this case, firms provide the best possible service given their infrastructure, but without an explicit guarantee. Using the Nash equilibria of the games played by firms, we compare this competitive model with the model where firms set prices and SLGs, in terms of the resulting prices, service levels, firms' profits, and consumers' surplus.

Traffic-Oblivious Routing for Guaranteed Bandwidth Performance

A virtual private network provides network connectivity among the different sites of an enterprise. The bandwidth requirement of a VPN is typically specified by peak ingress-egress traffic volumes for each site. The actual distribution of traffic from each site to other sites in the VPN is not known a priori and could vary over time (e.g., at an intra-day or daily level or due to special activities). This is captured by the hose traffic model. It is the responsibility of the Internet service provider to provision the VPN so that variable traffic subject to aggregate ingress-egress constraints can be transported with bandwidth guarantees. We describe a capacity-efficient, robust, and traffic-oblivious routing strategy, called two-phase routing, that allows preconfiguration of the VPN such that all traffic patterns permissible within the network's natural ingress-egress capacity constraints can be routed with bandwidth guarantees. The scheme routes traffic in two phases: traffic entering the VPN is sent from the source to a set of intermediate nodes in predetermined split ratios that depend on the intermediate nodes, and then from the intermediate nodes to the final destination. (The traffic split ratios can also be generalized to depend on the source and/or destination of traffic.) A significant benefit of the described schemes is that the network can be configured for completely static operation while providing service-level guarantees to widely varying VPN traffic. This is unlike previous approaches based on direct source-destination routing, which need dynamic adaptation to changing traffic conditions in order to provide service-level guarantees

Algorithmic aspects of bandwidth trading

We study algorithmic problems that are motivated by bandwidth trading in next-generation networks. Typically, bandwidth trading involves sellers (e.g., network operators) interested in selling bandwidth pipes that offer to buyers a guaranteed level of service for a specified time interval. The buyers (e.g., bandwidth brokers) are looking to procure bandwidth pipes to satisfy the reservation requests of end-users (e.g., Internet subscribers). Depending on what is available in the bandwidth exchange, the goal of a buyer is to either spend the least amount of money so as to satisfy all the reservations made by its customers, or to maximize its revenue from whatever reservations can be satisfied. We model this as a real-time nonpreemptive scheduling problem in which machine types correspond to bandwidth pipes and jobs correspond to end-user reservation requests. Each job specifies a time interval during which it must be processed, and a set of machine types on which it can be executed. If necessary, multiple machines of a given type may be allocated, but each must be paid for. Finally, each job has associated with it a revenue, which is realized if the job is scheduled on some machine. There are two versions of the problem that we consider. In the cost minimization version, the goal is to minimize the total cost incurred for scheduling all jobs, and in the revenue maximization version the goal is to maximize the revenue of the jobs that are scheduled for processing on a given set of machines. We consider several variants of the problems that arise in practical scenarios, and provide constant factor approximations.

Decongestants for clogged networks

Interconnection networks are a key element in a wide variety of systems: massive parallel processors, local and system area networks, clusters of PCs and workstations, and Internet Protocol routers. They are essential to high performance in the form of high-bandwidth communications, with low latency, "quality of service" (guaranteed service levels), efficient switching, and flexibility of network topology, as embodied in Myrinet, InfiniBand, Quadrics, Advanced Switching, and similar interconnects. But, despite all the advances that modem interconnects offer, congestion is a growing problem as "lossless" interconnection networksrdquo those that do not allow data packets to be discarded" come to the fore.