Showing papers on "Service level objective published in 2018"

Strategies for warranty service in a dual-channel supply chain with value-added service competition

Abstract: This paper studies a dual-channel supply chain composed of a manufacturer and a retailer. The manufacturer and the retailer sell homogeneous durable goods bundled with warranty service that is prov...

Exploring the innovation modes and evolution of the cloud-based service using the activity theory on the basis of big data

Abstract: With the cooperation between hospitals and advanced technologies such as mobile Internet, big data and cloud computing, emerging technologies and new service models have been applied in precaution, diagnosis, treatment and other medical services. As a result, people’s medical practices have been dramatically changed. On the basis of activity theory and service supply chain, the paper discusses hospital’s innovation in medical, health and care service by using cloud computing. It also explores the model and evolution of hospital’s internal activity and external supply chain.Cases in this paper from the year 2000 till now are our secondary data related to the cloud service innovation in hospitals. The study adopts textual analysis to interpret and analyze the data. Then, in-depth interviews would be taken on the basis of analytical result. Through the above-mentioned process, we find that homogeneities between different cases include treatment service cloud ranking first, community as the main promoter, substantial service innovation being emphasized, care service cloud as future trend, service innovation featured client-service model, eight procedures for service supply chain, improvement of suppliers’ self-design, and pluralism of medical care service etc. Differences of the cases include government policies versus industrial policies, multipoint distribution versus centralized organizational structure, regional institutionalization versus large-scale centralization, Substantive services application versus development of service value, market orientation versus industry orientation, service provider orientation versus service design orientation, physical customer versus invisible customer. In the end, the study offers relevant theories, management implications, and recommendations.

VCG Auction-Based Dynamic Pricing for Multigranularity Service Composition

Abstract: When a single service on its own cannot fulfill a sophisticated application, a composition of services is required. Existing methods mostly use a fixed-price scheme for service pricing and determine service allocation for composition based on a first-price auction. However, in a dynamic service market, it is difficult for service providers to determine a fixed price that is profitable while attractive to customers. Meanwhile, this mechanism cannot ensure that the providers who require the least cost to provide services would win the auction, because the pricing strategy of service providers is unpredictable. To address such issues, in this paper, we propose Vickrey–Clarke–Groves auction-based dynamic pricing for a generalized service composition. We consider fine-grained services as candidates for composition as well as coarse-grained ones. In our approach, service providers bid for services of different granularities in the composite service and based on received bids, a user decides a composition that minimizes the social cost while meeting quality constraints. Experimental results at last verify the feasibility and effectiveness of the proposed approach. Note to Practitioners —Motivated by the popular auction-based dynamic pricing in the modern Internet business, this paper proposes a Vickrey–Clarke–Groves auction-based approach for multigranularity service composition in order to overcome the shortcomings of existing approaches, e.g., unpredictability in service pricing, and lack of economic efficiency. The proposed approach instantiates a composite service in an economically efficient way and it offers service providers with the incentives to honestly declare the true costs of their concrete services in the bids. In the experiments, compared with fine-grained service composition, the proposed approach reduces the social cost and user payment of service composition by 6.8% and 11.2% on average, respectively, and it also shows superiority over the first-price auction-based approach.

Workforce management in omnichannel service centers with heterogeneous channel response urgencies

Abstract: Workforce staffing and assignment decisions are of critical importance for meeting the challenge of minimizing operational costs while providing satisfactory customer service. These decisions are particularly challenging for omnichannel service centers, where customers can request services via different communication channels (e.g., phone, e-mail, live-chat, social media) that have different service quality and response requirements. We present a formulation of the omnichannel workforce management problem that accounts for variations in response urgencies of different channels as well as diminishing agent performances due to channel switching. We develop an algorithm that efficiently provides solutions for this problem and determines the number and channel allocation of service agents within the service center. Through numerical experiments, we study the performance of the algorithm among various service center configurations with equal cost characteristics. The results indicate that the proposed algorithm can identify service center structures that outperform many alternative structures, including those commonly-adopted in the real-world.

Grus: Enabling Latency SLOs for GPU-Accelerated NFV Systems

Abstract: Graphics Processing Unit (GPU) has been recently exploited as a hardware accelerator to improve the performance of Network Function Virtualization (NFV). However, GPU-accelerated NFV systems suffer from significant latency variation when multiple network functions (NFs) are co-located in the same machine, which prevents operators from supporting latency Service Level Objectives (SLOs). Existing research efforts to address this problem can only guarantee a limited number of SLOs with very low resource utilization efficiency. In this paper, we present the Grus framework to support latency SLOs in GPU-accelerated NFV systems. Grus thoroughly analyzes the sources of latency variation and proposes three design principles: (1) dynamic batch size setting is needed to bound packet batching latency in CPU; (2) a reordering mechanism for data transfer over PCI-E is required to guarantee the stalling time; and (3) maximizing concurrency in GPU is necessary to avoid NF execution waiting time. Guided by the principles, Grus consists of two logical layers including an infrastructure layer and a scheduling layer. The infrastructure layer is equipped with an in-CPU Reorder-able Worker Pool that could adjust batching size and packet transfer order, and in-GPU Controllable Concurrent Executors to provide maximized concurrency. The scheduling layer runs a heuristic algorithm to perform accurate and fast scheduling to guarantee SLOs based on our prediction models. We have implemented a prototype of Grus. Extensive evaluations demonstrate that Grus can significantly reduce latency variation and satisfy 4.5× more SLO terms than state-of-the-art solutions.

Fast and Accurate Load Balancing for Geo-Distributed Storage Systems

Abstract: The increasing density of globally distributed datacenters reduces the network latency between neighboring datacenters and allows replicated services deployed across neighboring locations to share workload when necessary, without violating strict Service Level Objectives (SLOs). We present Kurma, a practical implementation of a fast and accurate load balancer for geo-distributed storage systems. At run-time, Kurma integrates network latency and service time distributions to accurately estimate the rate of SLO violations for requests redirected across geo-distributed datacenters. Using these estimates, Kurma solves a decentralized rate-based performance model enabling fast load balancing (in the order of seconds) while taming global SLO violations. We integrate Kurma with Cassandra, a popular storage system. Using real-world traces along with a geo-distributed deployment across Amazon EC2, we demonstrate Kurma's ability to effectively share load among datacenters while reducing SLO violations by up to a factor of 3 in high load settings or reducing the cost of running the service by up to 17%.

SLAs in 5G: A Complete Framework Facilitating VNF- and NS- Tailored SLAs Management

Abstract: As the 5G technology is expected to impact the mobile network and associated ecosystems, appropriate guarantees for the service quality can maximize the ability of Virtual Network Functions (VNF) and Network Services (NS). This implies the utilization of Service Level Agreements (SLA) to ensure that the NSs are provided in an efficient and controllable way. However, the complexity of determining resource provision policies in such multimodal environments as well as the characteristics and properties of various VNFs and NSs, results to custom SLAs that do not consider all aspects of the 5G environment. In this paper we propose an SLA Management Framework to map high-level requirements expressed by users, to low-level resource network parameters to improve the service provider's ability to meet the corresponding SLA commitments. In addition, we consider a mechanism for dynamic SLA Templates generation with initial Service Level Objectives (SLO) tailored to each service provider.

Henge: Intent-driven Multi-Tenant Stream Processing

Abstract: We present Henge, a system to support intent-based multi-tenancy in modern distributed stream processing systems. Henge supports multi-tenancy as a first-class citizen: everyone in an organization can now submit their stream processing jobs to a single, shared, consolidated cluster. Secondly, Henge allows each job to specify its own intents (i.e., requirements) as a Service Level Objective (SLO) that captures latency and/or throughput needs. In such an intent-driven multi-tenant cluster, the Henge scheduler adapts continually to meet jobs' respective SLOs in spite of limited cluster resources, and under dynamically varying workloads. SLOs are soft and are based on utility functions. Henge's overall goal is to maximize the total system utility achieved by all jobs in the system. Henge is integrated into Apache Storm and we present experimental results using both production jobs from Yahoo! and real datasets from Twitter.

5G & SLAs: Automated proposition and management of agreements towards QoS enforcement

Abstract: Efficient Service Level Agreements (SLA) management and anticipation of Service Level Objectives (SLO) breaches become mandatory to guarantee the required service quality in software- defined and 5G networks. To create an operational Network Service, it is highly envisaged to associate it with their network-related parameters that reflect the corresponding quality levels. These are included in policies but while SLAs target usually business users, there is a challenge for mechanisms that bridge this abstraction gap. In this paper, a generic black box approach is used to map high-level requirements expressed by users in SLAs to low-level network parameters included in policies, enabling Quality of Service (QoS) enforcement by triggering the required policies and manage the infrastructure accordingly. In addition, a mechanism for determining the importance of different QoS parameters is presented, mainly used for “relevant” QoS metrics recommendation in the SLA templates.

Service gap deployment: a framework to link quality gaps to service activities

Abstract: This paper presents the service gap deployment (SGD), a new method to prioritise crucial to quality activities of a service that does not completely satisfy customer expectations. In the SGD, service activities (SAs) are related to gaps between customer expectations and perceptions so as to identify SAs that may need a redesign or improvement in order to satisfy customer needs. The method, which expands the gaps model of service quality by Parasuraman et al. (1985), supports the design of evolutionary and sustaining improvements of the service parts that generate customer dissatisfaction. Specifically, the SGD introduces three major contributions: 1) it creates a map relating service dimensions to SAs; 2) it highlights crucial to quality activities; 3) it allows a focused improvement of the analysed service. The description is supported by an excerpt from a real application example, concerning the prioritisation of crucial to quality SAs of an airport luggage delivery service.

Revisiting Service Quality through the Lens of Experience-Centric Services

Abstract: The purpose of this research is to revisit prevailing notions of service quality by developing and testing a model of service quality for experience-centric services. By problematizing the service quality literature, a model is developed to capture impacts of outcome-achievement, instrumental performance and expressive performance on customer loyalty. A multi-group structural equation model is tested to establish the moderating effect of perceived service character—utilitarian or hedonic. Outcome-achievement mediates the direct relationships between instrumental and expressive performance, respectively, and loyalty; the strength of these relationships is moderated by perceived service character. Emotional design to improve the experience is effective provided the expected outcome is achieved. However, for services that customers perceive as experience-centric, the outcome may be somewhat ambiguously defined and expressive performance is valued more highly than instrumental performance. Understanding customers’ perception of a service—whether customers seek value related to outcomes or emotions—is crucial when selecting appropriate measures of service quality and performance. Creating a good experience is generally beneficial, but it must be designed according to the character of the service in question. The research presents empirical evidence on how service experience contributes to customer loyalty by testing a model of service quality that is suited to experience-centric services. Furthermore, it identifies the importance of understanding service character when designing and managing services.

Trust-based service composition and selection in service oriented architecture

Abstract: Service composition is a complex task that has attracted an increasing attention. It is significant to develop an approach to evaluate the trust of a composite service so that the client can obtain a satisfactory service from a large number of services. While, the fact is that many current approaches do not consider the trust dependency between component services in a composite service. Moreover, the approaches considering it focus only on the evaluation of already implemented composite service that can provide the rating of each component service in each execution. In this paper, we propose a method for evaluating the global trust of a composite service. It can be used for the evaluation of unexecuted composite services. In the trust evaluation procedure, we take trust dependency and the composite service’s invocation structure into account. We use a trust dependency analysis method inspired by the static program analysis technology to obtain the trust dependency relationship between component services. A greedy algorithm is also proposed to select the optimal component services from several candidate services for a composite service with the linear time complexity. Analysis and experiments show that our proposed approaches can evaluate the trust of a composite service reasonably and discover the optimal component service efficiently.

End-to-End Service Level Agreement Specification for IoT Applications

Abstract: The Internet of Things (IoT) promises to help solve a wide range of issues that relate to our wellbeing within applica¬tion domains that include smart cities, healthcare monitoring, and environmental monitoring. IoT is bringing new wireless sensor use cases by taking advantage of the computing power and flexibility provided by Edge and Cloud Computing. However, the software and hardware resources used within such applications must perform correctly and optimally. Especially in applications where a failure of resources can be critical. Service Level Agreements (SLA) where the performance requirements of such applications are defined, need to be specified in a standard way that reflects the end-to-end nature of IoT application domains, accounting for the Quality of Service (QoS) metrics within every layer including the Edge, Network Gateways, and Cloud. In this paper, we propose a conceptual model that captures the key entities of an SLA and their relationships, as a prior step for end-to-end SLA specification and composition. Service level objective (SLO) terms are also considered to express the QoS constraints. Moreover, we propose a new SLA grammar which considers workflow activities and the multi-layered nature of IoT applications. Accordingly, we develop a tool for SLA specification and composition that can be used as a template to generate SLAs in a machine-readable format. We demonstrate the effectiveness of the proposed specification language through a literature survey that includes an SLA language comparison analysis, and via reflecting the user satisfaction results of a usability study.

End-to-End Service Level Agreement Specification for IoT Applications

Abstract: The Internet of Things (IoT) promises to help solve a wide range of issues that relate to our wellbeing within application domains that include smart cities, healthcare monitoring, and environmental monitoring. IoT is bringing new wireless sensor use cases by taking advantage of the computing power and flexibility provided by Edge and Cloud Computing. However, the software and hardware resources used within such applications must perform correctly and optimally. Especially in applications where a failure of resources can be critical. Service Level Agreements (SLA) where the performance requirements of such applications are defined, need to be specified in a standard way that reflects the end-to-end nature of IoT application domains, accounting for the Quality of Service (QoS) metrics within every layer including the Edge, Network Gateways, and Cloud. In this paper, we propose a conceptual model that captures the key entities of an SLA and their relationships, as a prior step for end-to-end SLA specification and composition. Service level objective (SLO) terms are also considered to express the QoS constraints. Moreover, we propose a new SLA grammar which considers workflow activities and the multi-layered nature of IoT applications. Accordingly, we develop a tool for SLA specification and composition that can be used as a template to generate SLAs in a machine-readable format. We demonstrate the effectiveness of the proposed specification language through a literature survey that includes an SLA language comparison analysis, and via reflecting the user satisfaction results of a usability study.

FECBench: An Extensible Framework for Pinpointing Sources of Performance Interference in the Cloud-Edge Resource Spectrum

Abstract: Effective resource management is critical in multi-tenant, virtualized cloud platforms to meet service level objectives (SLOs) of individual applications. Thus, cloud providers must be able to detect sources of performance bottlenecks and reliability problems. One such cause, which is the focus of this study, is Performance Interference, where applications collocated on the same physical resource influence each others' performance in a non-linear fashion. In this paper, we present the challenges and requirements for an extensible performance interference benchmarking platform. This poster presents ongoing work on an extensible resource usage statistics collection and benchmarking framework we are developing called FECBench (Fog/Edge/Cloud Bench).

Customer differentiation with shipping as an ancillary service? free service, prioritization, and strategic delay

Abstract: A service provider/retailer offers ancillary service (e.g., shipping by an online retailer) to two types of customers, impatient and patient, who may be heterogeneous both in their delay sensitivities and service valuations. She can use prioritization and/or strategic delay to differentiate them by offering two service classes and charging different prices, potentially resulting in a split in which a single customer type selects both the classes. Her objective is to minimize cost while satisfying individual rationality and incentive compatibility conditions. We characterize the optimal solutions under both exogenous and endogenous capacities. We examine the conditions under which the following strategically important features of service delivery are optimal, and relate them to practical scenarios: (i) free service, (ii) single/differentiated service, (iii) split of customers, and (iv) strategic delay. We find that the presence of these features depends on (i) whether the retailer has limited or sufficient capacity and (ii) whether she sells fashion goods or staple products. A typical explanation for offering free service is that it increases demand from customers. We make an operational case for it by showing that even if demand does not change, free service is still optimal under some scenarios.

Less Provisioning: A Fine-grained Resource Scaling Engine for Long-running Services with Tail Latency Guarantees

Abstract: Modern resource management frameworks guarantee low tail latency for long-running services using the resource over-provisioning method, resulting in serious waste of resource and increasing the service costs greatly. To reduce the over-provisioning cost, we present EFRA, an elastic and fine-grained resource allocator that enables much more efficient resource provisioning while guaranteeing the tail latency Service Level Objective (SLO). EFRA achieves this through the cooperation of three key components running on a containerized platform: The period detector identifies the period features of the workload through a convolution-based time series analysis. The resource reservation component estimates the just-right amount of resources based on the period analysis through a top-K based collaborative filtering approach. The online reprovisioning component dynamically adjusts the resources for further enforcing the tail latency SLO. Testbed experiments show that EFRA is able to increase the average resource utilization to 43%, and save up to 66% resources while guaranteeing the same tail latency objective.

Robust Dynamic CPU Resource Provisioning in Virtualized Servers

Abstract: We present robust dynamic resource allocation mechanisms to allocate application resources meeting Service Level Objectives (SLOs) agreed between cloud providers and customers. In fact, two filter-based robust controllers, i.e. H-infinity filter and Maximum Correntropy Criterion Kalman filter (MCC-KF), are proposed. The controllers are self-adaptive, with process noise variances and covariances calculated using previous measurements within a time window. In the allocation process, a bounded client mean response time (mRT) is maintained. Both controllers are deployed and evaluated on an experimental testbed hosting the RUBiS (Rice University Bidding System) auction benchmark web site. The proposed controllers offer improved performance under abrupt workload changes, shown via rigorous comparison with current state-of-the-art. On our experimental setup, the Single-Input-Single-Output (SISO) controllers can operate on the same server where the resource allocation is performed; while Multi-Input-Multi-Output (MIMO) controllers are on a separate server where all the data are collected for decision making. SISO controllers take decisions not dependent to other system states (servers), albeit MIMO controllers are characterized by increased communication overhead and potential delays. While SISO controllers offer improved performance over MIMO ones, the latter enable a more informed decision making framework for resource allocation problem of multi-tier applications.

Service Level Objectives via C++11 Attributes

Abstract: In recent years, increasing attention has been given to the possibility of guaranteeing Service Level Objectives (SLOs) to users about their applications, either regarding performance or power consumption. SLO can be implemented for parallel applications since they can provide many control knobs (e.g., the number of threads to use, the clock frequency of the cores, etc.) to tune the performance and power consumption of the application. Different from most of the existing approaches, we target sequential stream processing applications by proposing a solution based on C++ annotations. The user specifies which parts of the code to parallelize and what type of requirements should be enforced on that part of the code. Our solution first automatically parallelizes the annotated code and then applies self-adaptation approaches at run-time to enforce the user-expressed objectives. We ran experiments on different real-world applications, showing its simplicity and effectiveness.

Data-driven resource flexing for network functions visualization

Abstract: Resource flexing is the notion of allocating resources on-demand as workload changes. This is a key advantage of Virtualized Network Functions (VNFs) over their non-virtualized counterparts. However, it is difficult to balance the timeliness and resource efficiency when making resource flexing decisions due to unpredictable workloads and complex VNF processing logic. In this work, we propose an Elastic resource flexing system for Network functions VIrtualization (ENVI) that leverages a combination of VNF-level features and infrastructure-level features to construct a neural-network-based scaling decision engine for generating timely scaling decisions. To adapt to dynamic workloads, we design a window-based rewinding mechanism to update the neural network with emerging workload patterns and make accurate decisions in real time. Our experimental results for real VNFs (IDS Suricata and caching proxy Squid) using workloads generated based on real-world traces, show that ENVI provisions significantly fewer (up to 26%) resources without violating service level objectives, compared to commonly used rule-based scaling policies.

Utility-based Allocation of Industrial IoT Applications in Mobile Edge Clouds

Abstract: Mobile Edge Clouds (MECs) create new opportunities and challenges in terms of scheduling and running applications that have a wide range of latency requirements, such as intelligent transportation systems, process automation, and smart grids. We propose a two-tier scheduler for allocating runtime resources to Industrial Internet of Things (IIoT) applications in MECs. The scheduler at the higher level runs periodically - monitors system state and the performance of applications - and decides whether to admit new applications and migrate existing applications. In contrast, the lower-level scheduler decides which application will get the runtime resource next. We use performance based metrics that tells the extent to which the runtimes are meeting the Service Level Objectives (SLOs) of the hosted applications. The Application Happiness metric is based on a single application's performance and SLOs. The Runtime Happiness metric is based on the Application Happiness of the applications the runtime is hosting. These metrics may be used for decision-making by the scheduler, rather than runtime utilization, for example. We evaluate four scheduling policies for the high-level scheduler and five for the low-level scheduler. The objective for the schedulers is to minimize cost while meeting the SLO of each application. The policies are evaluated with respect to the number of runtimes, the impact on the performance of applications and utilization of the runtimes. The results of our evaluation show that the high-level policy based on Runtime Happiness combined with the low-level policy based on Application Happiness outperforms other policies for the schedulers, including the bin packing and random strategies. In particular, our combined policy requires up to 30% fewer runtimes than the simple bin packing strategy and increases the runtime utilization up to 40% for the Edge Data Center (DC) in the scenarios we evaluated.

Monitoring application workloads scheduled on heterogeneous elements of information technology infrastructure

Abstract: A method includes selecting a given hardware configuration for a given application workload based on aligning an application workload specification template with a first hardware configuration template in a repository comprising a plurality of hardware configuration templates, the application workload specification template being generated by parsing and interpreting hardware-agnostic service level objective expressions of an application request. The method also includes scheduling the given application workload to run on information technology (IT) infrastructure utilizing the given hardware configuration, the given hardware configuration comprising a first set of a plurality of heterogeneous elements of the IT infrastructure, monitoring the IT infrastructure, and modifying the given hardware configuration for the given application workload based on aligning the application workload specification template with a second hardware configuration template in the repository responsive to said monitoring, the modified hardware configuration comprising a second set of the plurality of heterogeneous elements of the IT infrastructure.

Causes of Delivery-Time Variance in Maritime-Equipment Manufacturing Supply-Chains: An Empirical Study

Abstract: The overall performance of manufacturing companies has become increasingly dependent on their ability to coordinate a network of suppliers effectively. For manufacturers of customized equipment, it is even more important to coordinate several such network relationships concurrently to achieve service level objectives while minimizing inventory- and quality-related costs. In this paper, we investigate the causes of delivery variance in an engineer-to-order supply chain. Using four case companies within the global supply chain of a customized maritime-equipment manufacturer, we discuss these causes of delivery-time variance and suggestions for managing them.

Henge: Intent-driven Multi-Tenant Stream Processing

Abstract: We present Henge, a system to support intent-based multi-tenancy in modern stream processing applications Henge supports multi-tenancy as a first-class citizen: everyone inside an organization can now submit their stream processing jobs to a single, shared, consolidated cluster Additionally, Henge allows each tenant (job) to specify its own intents (ie, requirements) as a Service Level Objective (SLO) that captures latency and/or throughput In a multi-tenant cluster, the Henge scheduler adapts continually to meet jobs' SLOs in spite of limited cluster resources, and under dynamic input workloads SLOs are soft and are based on utility functions Henge continually tracks SLO satisfaction, and when jobs miss their SLOs, it wisely navigates the state space to perform resource allocations in real time, maximizing total system utility achieved by all jobs in the system Henge is integrated in Apache Storm and we present experimental results using both production topologies and real datasets

Mitigating write interference on SSD in home cloud server

Abstract: Recently, some home appliances such as smart TVs and home gateways adopt virtualization technologies to perform the role of home cloud server and govern all connected devices efficiently. In such virtualized systems, it is important to satisfy the I/O performance SLO (Service Level Objective) of each VM. In this paper, we propose a novel write buffer management scheme for SSDs in home cloud server, which guarantees the SLO of each VM by mitigating write interference problem among the VMs. Experimental results clearly show that our scheme outperforms the conventional write buffer scheme in terms of I/O performance SLO of each VM.

Elastic Services for Edge Computing

Abstract: Edge computing enables new, low-latency services close to data producers and consumers. However, edge service management is challenged by high hardware heterogeneity and missing elasticity capabilities. To address these challenges, this paper introduces the concept of elastic services. Elastic services are situation aware and can adapt themselves to the current execution environment dynamically to adhere to their Service Level Objectives (SLOs). This adaptation is achieved through Diversifiable Programming (DivProg), a new programming model which uses function annotations as interface between the service logic, its SLOs, and the execution framework. DivProg enables developers to characterize their services in a way that allows a third-party execution framework to run them with the flow and the parametrization that conforms to changing SLOs. We develop a prototype and perform an experimental evaluation which shows that elastic services can seamlessly adapt to heterogeneous platforms and scale with a wide range of input sizes, while adhering to their SLOs with little programming effort.

A Three-Dimensional Performance Measurement Model for Cloud Computing

Abstract: New paradigms for processing and storing data such as cloud computing require new approaches for the measurement of cloud service performance. To establish a Service Level Agreement (SLA) between a cloud service provider and its customers, the cloud services and their service level objectives need to be identified. An additional challenge in the performance measurement of cloud services is the lack of models that integrate the different perspectives of providers, maintainers and customers within the same model in order to define the concepts commonly used in cloud SLA contracts. This work proposes a three-dimensional Performance Measurement Model for Cloud Computing (P2M2C-3D) which consolidates performance measurement from the perspectives of providers, maintainers and customers for the different types of cloud services.

From Service to Service Design

Abstract: This chapter explores the origins of service design through an overview of key concepts and theories, starting from the definition of service and the characteristics of the service economy. Service marketing and service management literature are analyzed to describe the difference between products and services and to trace service peculiarities, which brought on the need of a dedicated design discipline and the formulation of the so-called Service-Dominant Logic, shifting the focus from services as goods to services as a perspective on value creation. The growing necessity and importance of designing services not only led to the birth of a new stream of study and practice in the field of design, through the development of specific methods and tools that support the creation of service solutions, but also allowed service design to become a crucial element for service innovation. Designing services that meet people’s and organizations’ needs, as well as the societal ones, is nowadays considered a strategic priority to support growth and prosperity. The final part of the chapter is therefore dedicated to outlining the role of service design in the contemporary socio-economic context as a driver for service, social and user-centered innovation.

A Multi-Agent Elasticity Management Based on Multi-Tenant Debt Exchanges

Abstract: A multi-tenant Software as a Service (SaaS) application is a highly configurable software that allows its owner to serve multiple tenants, each with their own workflows, workloads and Service Level Objectives (SLOs). Tenants are usually organizations that serve several users and the application appears to be a different one for each tenant. However, in practice, multi-tenant SaaS applications limit the diversity of tenants by clustering them in a few categories (e.g. premium, standard) with predefined SLOs. Additionally, this coarse-grained clustering reduces the advantage of these multi-tenant ecosystems over single tenant architectures to share dynamically virtual resources between tenants based on their own workload profile and elasticity adaptation decisions. To address this limitation, we propose a multi-agent elasticity management where each tenant is represented by a reinforcement learning agent that performs elasticity adaptations based on a new technical debt perspective, and make use of debt attributes (i.e. amnesty, interest) to form autonomous coalitions that minimise the effect of the unavoidable imperfections in any elasticity management approach. We extended CloudSim and Burlap to evaluate our approach. The simulation results indicate that our debt-aware multi-agent elasticity management preserves the diversity of tenants and reduces SLO violations without affecting the aggregate utility of the application owner.

Why is garbage collection causing my service level objectives to fail

Abstract: Cloud computing abstracts resources and provides them as-a-service to its tenant clients. Platform as a service clouds, which are one of the main types of cloud computing, provide large parts of the hardware/software stack to their users. Cloud systems are expected to abide by certain service level objectives and maintain a certain quality of service, which can be impacted by garbage collection (GC). However, cloud benchmarking is mostly focused in the interconnectivity of cloud services and often neglects the inner workings of language runtimes. In this paper, we present and evaluate CloudGC, a benchmark aiming to stress the GC component of a runtime in various and controllable ways. We then deploy our CloudGC on a cloud system to evaluate the SLO satisfaction of the four GC policies of the IBM J9 Java runtime. Our findings indicate that the default policy Gencon generally outperforms the other three policies, including Balanced, the policy which aims in amortising the costs.