Layered Queuing Model Research Articles

Bounds and limit theorems for a layered queueing model in electric vehicle charging

The rise of electric vehicles (EVs) is unstoppable due to factors such as the decreasing cost of batteries and various policy decisions. These vehicles need to be charged and will therefore cause congestion in local distribution grids in the future. Motivated by this, we consider a charging station with finitely many parking spaces, in which electric vehicles arrive in order to get charged. An EV has a random parking time and a random charging time. Both the charging rate per vehicle and the charging rate possible for the station are assumed to be limited. Thus, the charging rate of uncharged EVs depends on the number of cars charging simultaneously. This model leads to a layered queueing network in which parking spaces with EV chargers have a dual role, of a server (to cars) and a customer (to the grid). We are interested in the performance of the aforementioned model, focusing on the fraction of vehicles that get fully charged. To do so, we develop several bounds and asymptotic (fluid and diffusion) approximations for the vector process which describes the total number of EVs and the number of not fully charged EVs in the charging station, and we compare these bounds and approximations with numerical outcomes.

Tracking adaptive performance models using dynamic clustering of user classes (abstracts only)

Estimation techniques have been largely applied to track hidden performance parameters (e.g. service demands) of web based software systems. In this paper we investigate dynamic multiclass modeling of such systems, with variable classes of service, aiming at finding a low complexity model yet with enough accuracy. We propose a combination of clustering algorithm and tracking filter for effective grouping of classes of services. The tracking estimator is based on a layered queuing model with parameters for CPU demands and the user load intensity of each class of service. Clustering uses the K-means algorithm. The target application is autonomic control of web clusters, where changes occur at different rates and amplitudes and at random time instants. Experiments show that the tracking is effective, and reveal good filter settings for different variations.

Abstract only

Estimation techniques have been largely applied to track hidden performance parameters (e.g. service demands) of web based software systems. In this paper we investigate dynamic multiclass modeling of such systems, with variable classes of service, aiming at finding a low complexity model yet with enough accuracy. We propose a combination of clustering algorithm and tracking filter for effective grouping of classes of services. The tracking estimator is based on a layered queuing model with parameters for CPU demands and the user load intensity of each class of service. Clustering uses the K-means algorithm. The target application is autonomic control of web clusters, where changes occur at different rates and amplitudes and at random time instants. Experiments show that the tracking is effective, and reveal good filter settings for different variations.

Performance modeling of distributed multi-tier enterprise systems

Management techniques for achieving self-adaptive behavior in enterprise computing systems has been recently investigated by both academia and industry. Primarily, these techniques observe application measurements and take corrective action, based on a given model, to achieve a specified Quality of Service (QoS). Examples of such approaches include task scheduling, CPU provisioning [1], and power management [2]. Effective management of enterprise systems require finegrained decisions that balance various, often conflicting, service level agreements while adjusting to changes in the operating environment, caused by factors such as time-varying user workload and incomplete knowledge of the system operating state. Such decisions, affect how an application behaves under different workload intensities [3]. Over time, they influence system reliability, energy efficiency and customer experience. These decisions are implemented through a control system, requiring a detailed model of the system that reflects how various management choices affect the system behavior under varying environment inputs and operating conditions. An example of such a model is the layered queuing model presented in [4]. In this paper, we present the results of recent experiments aiming to model a distributed multi-tier enterprise application. Using dynamic regression and queuing modeling techniques, we have been able to obtain an approximate model structure that captures the system behavior under various operation conditions with a high degree of accuracy. In future, these models will be used to implement a distributed control structure that optimizes the system parameters for a given set of objectives [5]. This paper is organized as follows. Section 2 discusses the experimental setup. Section 3 discusses the performance management objectives, and discusses our experiments and modeling efforts. Section 4 concludes this paper.

An overhead and resource contention aware analytical model for overloaded Web servers

A Web server, when overloaded, shows a severe degradation of goodput initially, with the eventual settling of goodput as load increases further. Traditional performance models have failed to capture this behavior. In this paper, we propose an analytical model, which is a two-stage and layered queuing model of the Web server, which is able to reproduce this behavior. We do this by explicitly modelling the overhead processing, the user abandonment and retry behavior, and the contention for resources, for the FIFO and LIFO queuing disciplines. We show that LIFO provides better goodput in most overload situations. We compare our model predictions with experimental results from a test bed and find that our results match well with measurements.

Efficient performance models for layered server systems with replicated servers and parallel behaviour

Capacity planning for large computer systems may require very large performance models, which are difficult or slow to solve. Layered queueing models solved by mean value analysis can be scaled to dozens of servers and hundreds of service classes, with large class populations, but this may not be enough. A common feature of planning models for large systems is structural repetition expressed through replicated subsystems, which can provide both scalability and reliability, and this replication can be exploited to scale the solution technique. A model has recently been described for symmetrically replicated layered servers, and their integration into the system, with a mean-value solution approximation. However, parallelism is often combined with replication; high-availability systems use parallel data-update operations on redundant replicas, to enhance reliability, and grid systems use parallel computations for scalability. This work extends the replicated layered server model to systems with parallel execution paths. Different servers may be replicated to different degrees, with different relationships between them. The solution time is insensitive to the number of replicas of each replicated server, so systems with thousands or even millions of servers can be modelled efficiently.

Performance modeling and prediction of enterprise JavaBeans with layered queuing network templates

Component technologies, such as Enterprise Java Beans (EJB) and .NET, are used in enterprise servers with requirements for high performance and scalability. This work considers performance prediction from the design of an EJB system, based on the modular structure of an application server and the application components. It uses layered queueing models, which are naturally structured around the software components. This paper describes a framework for constructing such models, based on layered queue templates for EJBs, and for their inclusion in the server. The resulting model is calibrated and validated by comparison with an actual system.

Automated performance modeling of software generated by a design environment

Automation is needed to make performance modeling faster and more accessible to software designers. This paper describes a prototype tool that exploits recently developed techniques for automatic model construction from traces. The performance model is a layered queuing model and it is based on traces captured for certain selected scenarios which are determined to be important for performance. The prototype tool has been integrated with a commercial software design environment that generates code with heavy use of standard libraries. The execution costs of the libraries has also been captured and used in the automatic model creation. The approach not only automates building early models, but also gives models which can be maintained during development, using traces gathered from the implementations. The paper describes the tool, the process by which it is applied, the process of capturing and managing the execution cost data, and an example. To use the tool, the designer further defines the scenarios, the execution environment, and the workload intensity parameters. The designer can then experiment with different environments and workloads, or revise the design, to improve the performance.

Performance validation tools for software/hardware systems

It is common for software/hardware systems to be fully designed and functionally tested before an attempt is made to verify performance characteristics. Unsatisfactory performance, when discovered late in a system’s development, can cause a costly redesign and implementation of software or hardware and is likely to result in late system delivery. The goal of system performance validation is to provide assurance that a system as a whole is likely to meet its quantitative goals before the system is complete. It exploits performance engineering methods and tools to systematically construct and evaluate predictive system models. The tools must correctly predict system performance at some useful abstraction. This paper compares layered queueing models (LQMs) and stochastic process algebras (SPAs) and their support for system performance validation. In particular, we focus on abstraction and level detail within models and automated model building.

Performance Analytic Models and Analyses for Workflow Architectures

The design and implementation of a workflow management system is typically a large and complex task. Decisions need to be made about the hardware and software platforms, the data structures, the algorithms, and network interconnection of various modules utilized by various users and administrators. These decisions are further complicated by requirements such as flexibility, robustness, modifiability, availability, performance, and usability. As the size of workflow systems increases, organizations are finding that the standard server/client architectures, and off-the-shelf solutions are not adequate. We can further see that in the future, very large-scale workflow systems (VLSW) will become more complex, and more prevalent. Thus, one further requirement is an emphasis of this document: scalability. For the purposes of our scalable workflow investigations, we describe a framework, a taxonomy, a model, and a methodology to investigate the performance of various workflow architectures as the size of the system (number of workcases) grows very large. First, this paper presents a novel workflow architectural framework and taxonomy. We survey some example current workflow products and research prototype systems, illustrating some of the taxonomical categories. In fact, most current workflow architectures fall into only one of the many categories of this taxonomy: the centralized server/client category. The paper next explains a performance analysis methodology useful for exploring this taxonomy. The methodology deploys a layered queuing model, and performs mathematical analysis on this model using a modified MOL (method of layers) combined with a linearization algorithm. Finally, the paper utilizes this methodology to compare and contrast the various architectural categories, providing interesting results about performance as the number of workcases increases. Our analytic results suggest that (a) for VLSW performance determination, software architecture is as important as hardware architecture, and (b) alternatives to the client server architecture provide significantly better scalability.

Web server performance measurement and modeling techniques

The popularity of the Internet, and the usage of the world wide web in particular, has grown rapidly in recent years. Thousands of companies have deployed Web servers and their usage rates have increased dramatically. Our research has focused on measuring, analyzing and evaluating the performance of Internet and Intranet Web servers with a goal of creating capacity planning models. We have created layered queuing models (LQMs) and demonstrated their superiority to traditional queuing network models since they incorporate layered resource demands. Along the way we built a tool framework that enables us to collect and analyze the empirical data necessary to accomplish our goals. This paper describes the custom instrumentation we developed and deployed to collect workload metrics and model parameters from large-scale, commercial Internet and Intranet Web servers. We discuss the measurement issues pertaining to model parametrization and validation. We describe an object-oriented tool framework that significantly improves the productivity of analyzing the nearly 100 GBs of measurements collected during this workload study interval. Finally, we describe the LQM we developed to estimate client response time at a Web server. The model predicts the impact on server and client response times as a function of network topology and Web server pool size. We also use it to consider the consequences of server system configuration changes such as decreasing the HTTP object cache size.

Handover mechanism and speed considerations in personal communication systems

In this paper, we discuss views concerning different approaches to the analysis of personal communication systems (PCSs): an outside observer's point of view, a user's point of view, and a base station's point of view. In these three approaches the paper focuses on the handover mechanism rather than on the data traffic modeling. In particular, we reexamine these different methods of analysis, combine these views with our approach, and propose a new queueing model. We take into account the speed factor and show that the ordinary measurements that are based on the state probabilities are not appropriate. In addition, we propose a layered queueing model in which low-speed and high-speed users are treated differently.