Ubiquitous access to sophisticated internet services from diverse end devices across heterogeneous networks requires the injection of additional functionality into the network to handle protocol conversion, data transcoding, and in general bridge disparate network portions. Several researchers have proposed infrastructures for injecting such functionality; however, many challenges remain before these can be widely deployed. CANS is an application-level infrastructure for injecting application-specific components into the network that focuses on three such challenges: (a) efficient and dynamic composition of individual components; (b) distributed adaptation of injected components in response to system conditions; and (c) support for legacy applications and services. The CANS network view comprises applications, stateful services, and data paths built from mobile soft-state objects called drivers. Both services and data paths can be dynamically created and reconfigured: a planning and event propagation model assists in distributed adaptation, and a flexible type-based composition model dictates how new services and drivers are integrated with existing components. Legacy components plug into CANS using an interception layer that virtualizes network bindings and a delegation model. This paper describes the CANS architecture, and a case study involving a shrink-wrapped client application in a dynamically changing network environment where CANS improves overall user experience. 1

Future access to web-based content is likely to be dominated by two trends: (a) increasing amounts of dynamic, personalized content, and (b) a significant growth in “on-the-move” access using various mobile resource-constrained devices. These trends point to a situation where a user would have ubiquitous access to content, but require that content be efficiently delivered to the user irrespective of location, and in a form most suited to the user’s end device. Unfortunately, classical caching and transcoding solutions do not work well together, necessitating a new caching architecture built from the ground-up to handle problems caused by dynamic content, transcoded versions of objects, and the nomadic nature of users. This paper describes the goals and architecture of such a system: CONCA, an architecture for Consistent Nomadic Content Access.

Network aware application can achieve better performance by dynamically adapting to network service changes. The key question for network aware application development is how to obtain information about the performance of di erent system module. In this paper, we consider an important category of network aware application -- Compressed Data Transmission. Compression can reduce network transmission time by reducing the size of data to be transmitted, but on the other hand it increases local processing overhead. The tradeo between increased network processing and decreased local processing is critical to application's decision on how to transfer data. In this paper, we present our model to make such decision and discuss the methods of detecting network resources and predicting compression performance parameters. Experimental data on local testbed is presented to evaluate our methodology. We also discuss an improved model on how to deal with the overlap between network transmission and local processing, which has the potential to improve the application performance.

Applications on mobile computers must adapt to high variability in wireless network performance. Extending the semantics of transport protocols to offer more control over communication to the user allows applications to adapt their behavior to bandwidth variability. We examine adding bandwidth notifications, priorities and timeliness guarantees to a network API as a method for achieving greater application control over bursty traffic. Experiments demonstrate that the extended API allows applications to adjust to bandwidth variations effectively. We also compare three different implementations of the API: two which run on top of TCP, and one new protocol, ATP, which performs comparably to the TCP extensions, but has better performance for some workloads, including a workload simulating remote file system traffic.

A Web proxy must accurately predict network performance between itself and its servers and clients in order to make good distillation decisions. In this paper, we show that the current approaches to make such predictions --- either assuming the proxy is well-connected to all servers or using past observations --- are insufficient. We propose a new prediction method, estimation with uncertainty, that will play a crucial role in web proxies. This method can also be useful in domains such as distributed prefetching, distributed query planning, and cache replacement algorithms that take into account the cost of refetching evicted objects. 1. Introduction The diversity of speeds with which end users access network services is steadily increasing. Typical home users have network connectivity in the tens of Kb/s --- several orders of magnitude slower than institutional users. Mobile users can experience connectivity that frequently varies over similar ranges. Web services are often optimize...

A promising approach for providing seamless service access to portable and mobile end-devices is to augment the network path between client applications and services using “bridging ” components that are capable of caching, protocol conversion, transcoding, etc. While several such path-based approaches have been proposed, current approaches lack mechanisms for (1) automatically creating effective network paths whose performance is optimized for encountered network conditions, and (2) dynamically reconfiguring such paths when these conditions change. This paper describes our work on addressing these shortcomings. Our approach, which is built into an application-level programmable network infrastructure called CANS (Composable Adaptive Network Services), relies on modeling enhancements and algorithms to construct augmented network paths that not only improve application performance by coping with the resource gap between network services and clients, but can also dynamically adapt to changes in the network environment. We evaluate our approach over a range of network and end-device characteristics using two application scenarios: web access and image streaming. Our results validate the effectiveness of our approach for enabling network-aware service access to mobile clients, verifying that (1) data paths automatically created with our path creation algorithm do bring applications with considerable performance advantages; (2) fine tuned, desirable adaptation can be achieved using our flexible component model and reconfiguration strategy; and that (3) despite their flexibility, the run-time overhead of generated data paths is negligible, and the cost of path reconfiguration is small enough for most applications to continually adapt to dynamic changes. Key words: Adaptive middleware, Middleware for ubiquitous and mobile computing, Communication adaptation, Programmable networks. 1

Mobile Computing platforms such as mobile phones, PDAs or wearable computers operate in a much more volatile and limited environment than their stationary counterparts. Such platforms are inherently resource poor and subject to highly changeable resource availability. Applications for Mobile Computing require adaptation for best performance under such variable conditions, to make best use of available resources without assuming the minimum set. This paper details a framework developed by the authors for developing and deploying mobile applications. Current systems are able to notify an application to adapt but fail to say how. The application author must provide the actual adaptation mechanism. The authors aim to provide automatic adaptation to suitably constructed mobile applications. The adaptation mechanism is pervasive through application and system layers providing tight integration of adaptation both vertically (through an application) and horizontally (between applications). 1

Mobile devices must deal with limited and dynamically varying resources, in particular, the network quality of service (QoS). In addition, wireless devices have other constraints such as limited memory, battery power, and physical dimensions. Applications that execute in such environments need to adapt to the dynamic operating conditions in order to preserve an acceptable level of service as close to 100% of the time as possible. Viewing and downloading digital spatial data on mobile devices has become more popular, especially with the availability of "location-aware" applications that exploit GPS (Global Positioning System) receivers integrated in many of today's mobile devices. Map client applications face many challenges in accessing data across a wireless network. Vector spatial data files tend to be large, and file sizes tend to increase unpredictably depending on the complexity of feature geometry. Due to the limited size of the mobile device display, viewing all the details of the map could cause unreasonable clutter and render the map useless. Even if it is feasible to transmit all the details from a QoS standpoint, this could pose a problem from a usability standpoint. This research effort aims to tackle the issues of QoS and usability on mobile devices through a client-proxy-server model where clients are on mobile devices. The proxy performs two functions. First, it supplies the client with vital data about the status of the system that allows the client to take adaptive decisions aimed at maintaining the QoS. Second, it performs the adaptive actions requested by the client. There are two types of adaptive actions performed by the proxy, activating and deactivating filters. When filters are activated, the amount of data transmitted from the server to the clie...

Abstract not found

The Cadmium project provides system-level support for disconnected and mobile usage. This includes basic mechanisms for ensuring data availability, such as replication, caching, prefetching, as well as services for consistency, adaptation to environment changes, and so on. Cadmium uses an application-aware approach. It allows the applications or users, to be aware of currently available resources (e.g. network bandwidth). Application-awareness is achieved through cooperation between the system and the applications. The system provides environment monitoring, upcall registration and event notiøcation to send information about environment evolution. Applications dynamically adapt whenever required. Cadmium is still an on-going project.