Many modern networked applications require specific levels of service quality from the underlying network. Moreover, next-generation networked applications are expected to adapt to changes in the underlying network, services, and user interactions. While some applications have built-in adaptivity, the adaptation itself requires specification of a system model. This paper presents Sapphire, an experimental approach for systematic model generation for application adaptation within a target network. It employs a nearly-automated, statistical design of experiments to characterize the relationships of both application and network-level parameters. First, it applies the Analysis of Variance (ANOVA) method to identify the most significant parameters and their interactions that affect performance. Next, it generates a model of application performance with respect to these parameters within the ranges of measurements. The key benefit of the framework is the integration of several well-established concepts of statistical modeling and distributed systems in the form of simple APIs so that existing applications can take advantage of it. We demonstrate the usefulness and flexibility of Sapphire by generating a performance model of an audio streaming application. We show that many existing multimedia and QoS-sensitive applications can exploit a statistical modeling approach such as Sapphire to incorporate application adaptivity. The approach can also be used for feedback control of distributed applications, tuning network and application parameters to achieve service levels in a target network.

Access links are typically the bottleneck between a high bandwidth LAN and a high bandwidth IP network. Without a priori resource provisioning or reservation, this tends to have a serious effect on premium traffic from delay- and jitter-sensitive applications. This paper develops a new reservation-less mechanism for delay and jitter guarantees across "single FIFO queue" access links by controlling interfering non-premium traffic. The mechanism, called CONNET (CONtrolled NETwork), uses feedback regulation to create self-controlled network services capable of tracking delay and jitter references. We use a control-theoretic approach in designing the feedback loop based on a "black-box" model of the access link in both cases of single and multiple nodes. For robust control, we employ optimal Linear Quadratic Gaussian (LQG) regulator. We evaluate the performance of CONNET using a real testbed implementation showing its superiority to other rate-based schemes (such as CBQ and WFQ) in terms of simplicity, deployability, and accuracy.

Providing real-time communication services to multimedia applications and subscription-based Internet access often requires that sufficient network resources be reserved for real-time traffic. However, the reserved network resource is susceptible to resource theft and abuse. Without a resource access control mechanism that can efficiently differentiate legitimate real-time traffic from attacking packets, the traffic conditioning and policing enforced at Internet Service Provider (ISP) edge routers cannot protect the reserved network resource from embezzlement. On the contrary to the usual expectation, the traffic policing at edge routers aggravates their vulnerability to flooding attacks by blindly dropping packets. In this paper, we propose a fast and lightweight IP network-edge resource access control mechanism, called IP Easy-pass, to prevent unauthorized access to reserved network resources at edge devices. We attach a unique pass to each legitimate real-time packet so that an ISP edge router can validate the legitimacy of the incoming IP packet very quickly and simply by checking its pass. We present the generation of Easy-pass, its embedding, and verification procedures. We implement the IP Easy-pass mechanism in the Linux kernel, and measure its overhead on our testbed. Finally, we demonstrate its effectiveness against packet forgery and resource embezzlement attempts by conducting a series of experiments.

Providing real-time communication services to multimedia applications and subscription-based Internet access often requires sufficient network resources to be reserved for real-time traffic. However, the reserved network resource is susceptible to resource theft and abuse. Without a resource access control mechanism that can efficiently differentiate legitimate real-time traffic from attacking packets, the traffic conditioning and policing enforced at ISP (Internet service provider) edge routers cannot protect the reserved network resource from embezzlement. On the contrary, the traffic policing at edge routers aggravates their vulnerability to flooding attacks by blindly dropping packets. We propose a fast and light-weighted IP network-edge resource access control mechanism, called IP easy-pass to prevent unauthorized access to reserved network resources at edge devices. We attach a unique pass to each legitimate real-time packet so that an ISP edge router can validate the legitimacy of an incoming IP packet very quickly and simply by checking its pass. We present the generation of easy-pass, its embedding, and verification procedures. We implement the IP easy-pass mechanism in the Linux kernel, analyze its effectiveness against packet forgery and resource embezzlement attempts. Finally, we measure the overhead incurred by easy-pass.

Paving the first mile of quality-of-service (QoS) support has become essential for full deployment and utilization of QoS proposals in the Internet. Most of existing efforts have been made to provide network services and control without paying much attention to how applications can use these services. In this paper, we design and implement a two-tier architecture, called QoS Gateway (QoSGW), that acts as an interface between application QoS requirements and network-provided QoS capabilities. The QoSGW is to support small embedded network devices that rely on network-provided QoS. Our architecture, in its full version, is composed of two key components: (i) an agent that resides on the end-host and provides an adequate interface for QoS-dependent applications, and (ii) a QoS manager that provides an interface to network services for the agents. Using these two components enhances generality and scalability in providing QoS support for Internet applications and end-devices. The QoSGW is intended to promote QoS deployment and facilitate construction of QoS-aware access networks.

The past few years have witnessed the emergence of many real-time networked applications on the Internet. These types of applications require special support from the underlying network such as reliability, timeliness, and guaranteed delivery, as well as different levels of service quality. Unfortunately, this support is not available within the current "best-effort" Internet architecture. In this paper, we review several mechanisms and frameworks proposed to provide network- and application-level quality of service (QoS) in the next-generation Internet. We first discuss the QoS requirements of many of the above-mentioned real-time applications, and then we categorize them according to the required service levels. We also describe the various building blocks often used in QoS approaches. We briefly present asynchronous transfer mode (ATM) and Internet Protocol precedence. Then, we present and compare two service architectures recently adopted by the Internet Engineering Task Force, called integrated services (IntServ) and differentiated services (DiffServ), for providing per-flow and aggregated-flow service guarantees, respectively. We focus on DiffServ because it is a candidate QoS framework to be used in next-generation Internet along with multiprotocol label switching and traffic engineering. We also examine several operational and research issues that need to be resolved before such frameworks can be put in practice.

This paper introduces a new packet marking algorithm that can be used in the context of Assured Forwarding (AF) in the Differentiated Services (DiffServ) framework. The new marking algorithm is called equation-based marking (EBM) and is based on the TCP model in Padhye et al. (1998). EBM is to handle the problems found in other marking schemes regarding fairness among heterogeneous TCP flows through a tight feedback-loop operation and adaptation of the packet marking probability to network conditions. We design a packet marker that uses EBM as the marking algorithm, and evaluate its performance using in-depth simulation. We also prove analytically the correctness of the marking algorithm and compare it with other marking schemes for different network scenarios. Our evaluation results demonstrate the effectiveness of EBM in providing the required fairness among heterogeneous flows and ensuring protection against nonassured traffic.