Resource Allocation for Session-Based Two-Dimensional Service Differentiation on e-Commerce Servers

A scalable e-commerce server should be able to provide different levels of quality of service (QoS) to different types of requests based on clients' navigation patterns and the server capacity. E-Commerce workloads are composed of sessions. In this paper, we propose a session-based two-dimensional (2D) service differentiation model for online transactions: intersession and intrasession. The intersession model aims to provide different levels of QoS to sessions from different customer classes, and the intrasession model aims to provide different levels of QoS to requests in different states of a session. A primary performance metric of online transactions is slowdown. It measures the waiting time of a request relative to its service time. We present a processing rate allocation scheme for 2D proportional slowdown differentiation. We then introduce service slowdown as a systemwide QoS metric of an e-commerce server. It is defined as the weighted sum of request slowdown in different sessions and in different session states. We formulate the problem of 2D service differentiation as an optimization of processing rate allocation with the objective of minimizing the service slowdown of the server. We prove that the derived rate allocation scheme based on the optimization guarantees client requests' slowdown to be square-root proportional to their prespecified differentiation weights in both intersession and intrasession dimensions. We evaluate this square-root proportional rate allocation scheme and a proportional rate allocation scheme via extensive simulations. Results validate that both schemes can achieve predictable, controllable, and fair 2D service differentiation on e-commerce servers. The square-root proportional rate allocation scheme provides 2D service differentiation at a minimum cost of service slowdown.     

Harmonic proportional bandwidth allocation and scheduling for service differentiation on streaming servers

To provide ubiquitous access to the proliferating rich media on the Internet, scalable streaming servers must be able to provide differentiated services to various client requests. Recent advances of transcoding technology make network-I/O bandwidth usages at the server communication ports controllable by request schedulers on the fly. In this article, we propose a transcoding-enabled bandwidth allocation scheme for service differentiation on streaming servers. It aims to deliver high bit rate streams to high priority request classes without overcompromising low priority request classes. We investigate the problem of providing differentiated streaming services at application level in two aspects: stream bandwidth allocation and request scheduling. We formulate the bandwidth allocation problem as an optimization of a harmonic utility function of the stream quality factors and derive the optimal streaming bit rates for requests of different classes under various server load conditions. We prove that the optimal allocation, referred to as harmonic proportional allocation, not only maximizes the system utility function, but also guarantees proportional fair sharing between classes with different prespecified differentiation weights. We evaluate the allocation scheme, in combination with two popular request scheduling approaches, via extensive simulations and compare it with an absolute differentiation strategy and a proportional-share strategy tailored from relative differentiation in networking. Simulation results show that the harmonic proportional allocation scheme can meet the objective of relative differentiation in both short and long timescales and greatly enhance the service availability and maintain low queueing delay when the streaming system is highly loaded.     

Regression-based resource provisioning for session slowdown guarantee in multi-tier Internet servers

Autonomous management of a multi-tier Internet service involves two critical and challenging tasks, one understanding its dynamic behaviors when subjected to dynamic workloads and second adaptive management of its resources to achieve performance guarantees. We propose a statistical machine learning based approach to achieve session slowdown guarantees of a multi-tier Internet service. Session slowdown is the relative ratio of a session’s total queueing delay to its total processing time. It is a compelling performance metric of session-based online transactions because it directly measures user-perceived relative performance and it is independent of the session length. However, there is no analytical model for session slowdown on multi-tier servers. We first conduct training to learn the statistical regression models that quantitatively capture an Internet service’s dynamic behaviors as relationships between various service parameters. Then, we propose a dynamic resource provisioning approach that utilizes the learned regression models to efficiently achieve session slowdown guarantee under dynamic workloads. The approach is based on the combination of offline training and online monitoring of the Internet service behavior. Simulations using the industry standard TPC-W benchmark demonstrate the effectiveness and efficiency of the regression based resource provisioning approach for session slowdown oriented performance guarantee of a multi-tier e-commerce application.     

Web服务器中基于自适应控制的成比例延迟区分服务

Internet上Web应用日益广泛的使用,使得Web服务器需要在高负载下提供性能保证与区分服务,以满足用户的不同需求。响应延迟是评价Web服务器的一项关键性能指标,而成比例延迟区分服务是一种重要的区分服务模型。针对Apache Web服务器,提出并实现了基于自适应控制的成比例延迟区分服务。在每个采样周期,自适应控制器根据预设的延迟区分参数,通过动态计算并调节各个客户类别的服务线程数目,可保证Apache Web服务器上高优先级客户具有较低的平均连接延迟,而各个客户类别的平均延迟比保持不变。仿真结果表明,在动态变化的负载、参考输入以及不同的系统配置之下,控制器作用下的Apache Web服务器都能可靠地提供成比例延迟区分服务。     

PAPMSC: Power-Aware Performance Management Approach for Virtualized Web Servers via Stochastic Control

As green computing is becoming a popular computing paradigm, the performance of energy-efficient data center becomes increasingly important. This paper proposes power-aware performance management via stochastic control method (PAPMSC), a novel stochastic control approach for virtualized web servers. It addresses the instability and inefficiency issues due to dynamic web workloads. It features a coordinated control architecture that optimizes the resource allocation and minimizes the overall power consumption while guaranteeing the service level agreements (SLAs). More specifically, due to the interference effect among the co-located virtualized web servers and time-varying workloads, the relationship between the hardware resource assignment to different virtual servers and the web applications’ performance is considered as a coupled Multi-Input-Multi-Output (MIMO) system and formulated as a robust optimization problem. We propose a constrained stochastic linear-quadratic controller (cSLQC) to solve the problem by minimizing the quadratic cost function subject to constraints on resource allocation and applications’ performance. Furthermore, a proportional controller is integrated to enhance system stability. In the second layer, we dynamically manipulate the physical frequency for power efficiency using an adaptive linear quadratic regulator (ALQR). Experiments on our testbed server with a variety of workload patterns demonstrate that the proposed control solution significantly outperforms existing solutions in terms of effectiveness and robustness.     

A novel resource allocation mechanism for live cloud-based video streaming service

With the recent emergence of cloud computing, growing numbers of clients are using online cloud services through the Internet such as video streaming service. The rent costs of cloud service providers increase when the resource utilizations of the cloud-servers are not well. Therefore, resource allocation is a crucial problem for cloud data centers. The resource allocation problem is an NP-hard problem. This paper proposes a novel cloud resource allocation mechanism based on a winning strategy for a Nim game. This mechanism offers all clients an effective number of running cloud servers, and allocates cloud resources rapidly and effectively by using a pre-pairing approach. The proposed mechanism does not require searching for remaining resources of the running cloud server; hence, it can reduce the time taken to arrange resources. The experimental results show that the proposed mechanism can improve utilization of cloud servers and reduce the rent costs of the cloud service providers. The proposed mechanism can reach the utilization of cloud servers by as much as 99.96 %. The proposed mechanism is approximately 9 % more efficient than the market-based grid resource allocation algorithm, and 19 % more efficient than the modified best fit decreasing algorithm.

     

Policy-based Diffserv on Internet servers: the AIX approach

The Internet infrastructure must evolve from best-effort service to meet the needs of different customers and applications. With Internet traffic differentiation, service providers can support a range of offerings, such as loss or delay bounds and network bandwidth allocation, to meet different performance requirements. The differentiated services (Diffserv) architecture provides a scalable approach, in which network access (or edge) devices aggregate traffic flows onto provisioned pipes that traverse a streamlined network core. We have identified the key requirements for provisioning Diffserv functions on Internet servers. Based on these requirements, we have implemented, and deployed, a policy-based architecture on IBM's AIX operating system that provides Diffserv services to both QoS-aware and -unaware applications     

采用随机规划模型的云资源分配算法

针对云资源提供问题,为了降低云消费者的资源使用成本,提出了一种采用随机规划模型的云资源分配算法.同时考虑按需实例和预留实例,采用两阶段随机整数规划对云资源提供问题进行建模,在资源预留阶段,根据长期的工作负载情况,确定预留实例的类型和数量,在按需分配阶段,根据当前的工作负载,确定动态分配的按需实例的类型和数量.采用抽样平均近似方法减少资源提供问题的场景数量,降低求解复杂度,并提出了一种基于阶段分解的混合进化算法求解资源提供问题.仿真实验结果表明,采用随机规划模型的云资源分配算法能够在较短时间内获得近似最优的云资源预留方案,有效降低了云消费者的资源使用成本.     

移动边缘计算中资源受限的动态服务部署策略

移动边缘计算（MEC）的出现使移动用户能够以低延迟访问部署在边缘服务器上的服务。然而,MEC仍然存在各种挑战,尤其是服务部署问题。边缘服务器的数量和资源通常是有限的,只能部署数量有限的服务;此外,用户的移动性改变了不同服务在不同地区的流行度。在这种情况下,为动态请求部署合适的服务就成为一个关键问题。针对该问题,通过了解动态用户请求来部署适当的服务以最小化交互延迟,将服务部署问题表述为一个全局优化问题,并提出了一种基于集群划分的资源聚合算法,从而在计算、带宽等资源约束下初步部署合适的服务。此外,考虑动态用户请求对服务流行度及边缘服务器负载的影响,开发了动态调整算法来更新现有服务,以确保服务质量（QoS）始终满足用户期望。通过一系列仿真实验验证了所提出策略的性能。仿真结果表明,与现有基准算法相比,所提出的策略可以降低服务交互延迟并实现更稳定的负载均衡。     

DTN Web服务器基于自适应控制的成比例延迟区分服务

DTN网络的出现为未来的Internet应用提供了一种全新的平台,在DTN网络上部署支持服务区分服务的Web服务器方面的研究仍为空白。首先设计了一种支持区分服务的DTN Web服务器,并在其上实现了基于自适应控制的成比例延迟区分服务。自适应控制器根据预设的延迟区分参数,通过动态计算并调节各个客户类别的服务线程数目,可保证DTN Web服务器上高优先级客户具有较低的平均连接延迟而各个客户类别的平均延迟比保持不变。仿真结果表明,在动态变化的负载、不同的负载分布以及变化的参考输入之下,控制器作用下的DTN Web服务器都能可靠地提供成比例延迟区分服务。     

Resource consumption-aware QoS in cluster-based VOD servers

For Video-On-Demand (VOD) systems, it is important to provide Quality of Service (QoS) to more clients under limited resources. In this paper, the performance scalability in cluster-based VOD servers is studied with several grouping configurations of cluster nodes. To find performance bottlenecks, the monitoring functions are employed and the maximum QoS streams are measured under the various requests including VCR functions. To support more user friendly interface, an embedded set-top model is suggested for the QoS of TV clients. From our detailed experiment results, a new admission control method is proposed that is based on available system resources and the actual amount of resource consumed for QoS streams. The proposed method provides not only more scalable QoS in cluster-based VOD servers but also the enhancement of resource utilization by guaranteeing the maximum number of QoS streams.     

Dynamic resource demand prediction and allocation in multi‐tenant service clouds

Cloud computing is emerging as an increasingly popular computing paradigm, allowing dynamic scaling of resources available to users as needed. This requires a highly accurate demand prediction and resource allocation methodology that can provision resources in advance, thereby minimizing the virtual machine downtime required for resource provisioning. In this paper, we present a dynamic resource demand prediction and allocation framework in multi‐tenant service clouds. The novel contribution of our proposed framework is that it classifies the service tenants as per whether their resource requirements would increase or not; based on this classification, our framework prioritizes prediction for those service tenants in which resource demand would increase, thereby minimizing the time needed for prediction. Furthermore, our approach adds the service tenants to matched virtual machines and allocates the virtual machines to physical host machines using a best‐fit heuristic approach. Performance results demonstrate how our best‐fit heuristic approach could efficiently allocate virtual machines to hosts so that the hosts are utilized to their fullest capacity. Copyright © 2016 John Wiley & Sons, Ltd.     

Minimizing Resource Cost for Camera Stream Scheduling in Video Data Center

Video surveillance service, which receives live streams from IP cameras and forwards the streams to end users, has become one of the most popular services of video data center. The video data center focuses on minimizing the resource cost during resource provisioning for the service. However, little of the previous work comprehensively considers the bandwidth cost optimization of both upload and forwarding streams, and the capacity of the media server. In this paper, we propose an efficient resource scheduling approach for online multi-camera video forwarding, which tries to optimize the resource sharing of media servers and the networks together. Firstly, we not only provide a fine-grained resource usage model for media servers, but also evaluate the bandwidth cost of both upload and forwarding streams. Without loss of generality, we utilize two resource pricing models with different resource cost functions to evaluate the resource cost: the linear cost function and the non-linear cost functions. Then, we formulate the cost minimization problem as a constrained integer programming problem. For the linear resource cost function, the drift-plus-penalty optimization method is exploited in our approach. For non-linear resource cost functions, the approach employs a heuristic method to reduce both media server cost and bandwidth cost. The experimental results demonstrate that our approach obviously reduces the total resource costs on both media servers and networks simultaneously.     

Resource optimization algorithms for virtual private networks using the hose model

Virtual private networks (VPNs) provide a secure and reliable communication between customer sites over a shared network. With increase in number and size of VPNs, service providers need efficient provisioning techniques that adapt to customer demands. The recently proposed hose model for VPN alleviates the scalability problem of the pipe model by reserving for aggregate ingress and egress bandwidths instead of between every pair of VPN endpoints. Existing studies on quality of service guarantees in the hose model either deal only with bandwidth requirements or regard the delay limit as the main objective ignoring the bandwidth cost. In this work we propose a new approach to enhance the hose model to guarantee delay limits between endpoints while optimizing the provisioning cost. We connect VPN endpoints using a tree structure and our algorithm attempts to optimize the total bandwidth reserved on edges of the VPN tree. Further, we introduce a fast and efficient algorithm in finding the shared VPN tree to reduce the total provisioning cost compared to the results proposed in previous works. Our proposed approach takes into account the user preferences in meeting the delay limits and provisioning cost to find the optimal solution of resource allocation problem. Our simulation results indicate that the VPN trees constructed by our proposed algorithm meet maximum end-to-end delay limits while reducing the bandwidth requirements as compared to previously proposed algorithms.     

A cost‐efficient resource provisioning algorithm for DHT‐based cloud storage systems

Personal cloud storage provides users with convenient data access services. Service providers build distributed storage systems by utilizing cloud resources with distributed hash table (DHT), so as to enhance system scalability. Efficient resource provisioning could not only guarantee service performance, but help providers to save cost. However, the interactions among servers in a DHT‐based cloud storage system depend on the routing process, which makes its execution logic more complicated than traditional multi‐tier applications. In addition, production data centers often comprise heterogeneous machines with different capacities. Few studies have fully considered the heterogeneity of cloud resources, which brings new challenges to resource provisioning. To address these challenges, this paper presents a novel resource provisioning model for service providers. The model utilizes queuing network for analysis of both service performance and cost estimation. Then, the problem is defined as a cost optimization with performance constraints. We propose a cost‐efficient algorithm to decompose the original problem into a sub‐optimization one. Furthermore, we implement a prototype system on top of an infrastructure platform built with OpenStack. It has been deployed in our campus network. Based on real‐world traces collected from our system and Dropbox, we validate the efficiency of our proposed algorithms by extensive experiments. Copyright © 2016 John Wiley & Sons, Ltd.     

An analytical approach to providing controllable differentiated quality of service in Web servers

Provisioning quality of service (QoS) in Web servers has gained immense importance because Web servers are a major part of the Internet. To deliver the pledged QoS, Web service providers need control over the allocation of the resources in their Web servers. Control is also necessary for reaching the optimal resource allocation through proper service differentiation. In this paper, we propose and investigate an analytic approach that enables the service providers to deploy a differentiated service policy that offers this control. The proposed service policy is configurable by tunable control parameters. We devise the relationships between the performance measures and these parameters by adopting a unique queuing theoretic approach. Once these relationships are established, we describe how these parameters can be set to their most appropriate values depending on the objectives of the service providers. We illustrate the usefulness of our approach by conducting the analysis on a real Web trace.     

Quality-of-service differentiation on the Internet: A taxonomy

Internet applications and clients have very diverse service expectations, demanding for provisioning of different levels of quality of service (QoS) to multiple traffic classes on the Internet. To meet this demand, many different approaches and performance metrics have been proposed in an attempt to achieve per-class QoS differentiation in the network core, network edges, proxy and end servers. However, due to the wide variety of approaches to the problem, it is difficult to meaningfully compare different approaches since there is no uniform means for quantitatively or qualitatively evaluating them. Thus, it is hard to build upon existing work or identify areas worthy of additional efforts without understanding of the relationships between existing efforts.In this paper, a taxonomy of approaches to per-class QoS differentiation is presented in an attempt to provide a common terminology and classification mechanism necessary in addressing this problem. The taxonomy categorizes state-of-the-art QoS differentiation approaches in three dimensions. The first categorization dimension is the locations where the approaches are deployed, i.e., server, proxy and network sides. The QoS differentiation approaches are secondly categorized according to their policies, i.e., admission control, resource management, and content adaptation. The third categorization dimension is their implementation layer, i.e., application level, and kernel level. Representative QoS differentiation approaches in each category are reviewed.     

Web服务器系统中基于反馈控制的比例延迟保证

随着Web应用在商业领域的广泛使用,Web服务器系统需要在高负载下提供区分服务,以满足用户的不同需求。为实现以延迟作为评价指标的区分服务,本文在Web服务器系统的连接管理和请求处理两个层次建立了基于反馈控制的比例延迟保证模型。模型中的反馈控制器通过动态计算和调节不同类别客户占用的资源：（服务线程和数据库连接）,能保证高优先级的客户较快得到服务而不同类别客户的平均延迟比保持不变。为测试闭环系统的性能,设计了两种分别服从均匀分布和重尾分布的动态负载。仿真结果表明,即使并发客户连接的数目剧烈变化,控制器作用下的服务器系统仍然能够达到较好的比例延迟保证,可靠地为用户提供区分服务。     

Multi-Layer Resource Management in Cloud Computing

The paper studies multi-layer optimization in service oriented cloud computing to optimize the utility function of cloud computing, subject to resource constraints of an IaaS provider at the resource layer, service provisioning constraints of a SaaS provider at the service layer, and user QoS (quality of service) constraints of cloud users at application layer, respectively. The multi-layer optimization problem can be decomposed into three subproblems: cloud computing resource allocation problem, SaaS service provisioning problem, and user QoS maximization problem. The proposed algorithm decomposes the global optimization problem of cloud computing into three sub-problems via an iterative algorithm. The experiments are conducted to test the efficiency of the proposed algorithm with varying environmental parameters. The experiments also compare the performance of the proposed approach with other related work.     

Dynamic provisioning in multi-tenant service clouds

Cloud-based systems promise an on-demand service provisioning system along with a ??pay-as-you-use?? policy. In the case of multi-tenant systems this would mean dynamic creation of a tenant by integrating existing cloud-based services on the fly. Presently, dynamic creation of a tenant is handled by building the required components from scratch. Although multi-tenant systems help providers save cost by allocating multiple tenants to the same instance of an application, they incur huge reconfiguration costs. Cost and time spent on these reconfiguration activities can be reduced by re-constructing tenants from existing tenant configurations supported by service providers. Multi-tenant cloud-based systems also lack the facility of allowing clients to specify their requirements. Giving clients the flexibility to specify requirements helps them avoid spending an excessive amount of time and effort looking through a list of services, many of which might not be relevant to them. Moreover, dynamic provisioning in the cloud requires an integrated solution across the technology stack (software, platform and infrastructure) combining functional, non-functional and resource allocation requirements. Existing research works in the area of web service matching, although numerous, still fall short, since they usually consider each requirement type in isolation and cannot provide an integrated solution. To that end, in this paper we investigate the features needed for dynamic service provisioning on the cloud. We propose a novel User Interface-Tenant Selector-Customizer (UTC) model and approach, which enables cloud-based services to be systematically modeled and provisioned as variants of existing service tenants in the cloud. Our approach considers functional, non-functional and resource allocation requirements, which are explicitly specified by the client via the user interface component of the model. To the best of our knowledge, ours is the first such integrated approach. We illustrate our ideas using a realistic running example, and also present a proof-of-concept prototype built using IBM??s Rational Software Architect modeling tool. We also present experimental results demonstrating the applicability of our matching algorithm. Our results show significant reduction in matching time with the help of an elimination process that reduces the search space needed for performing matching.