Search

Nomadicity is a new challenge for computing and communication technologies. Modern cellular telephone systems extend the usability of portable personal computers enormously. A nomadic user can be given ubiquitous access to remote information stores and computing services. However, the behaviour of wireless links creates severe inconveniences within the traditional data communication paradigm. In this paper we give an overview of the problems related to wireless mobility. We also present a new software architecture for mastering the problems and discuss a new paradigm for designing mobile distributed applications. The key idea in the architecture is to place a mediator, a distributed intelligent agent, between the mobile node and the wireline network.

Modern digital cellular telephones and portable computers have created a new platform for distributed information processing. However, the characteristics of wireless telephone links are different from those of wireline links. With standard TCP/IP protocols, this can lead to severe performance problems; some are related to the control of the wireless link, some to the cooperation of the wireless link and the fixed network. One possible solution is to split the end-to-end communication path into two parts, and to establish a separate control for each part. The Mowgli communication architecture is a sophisticated elaboration of this basic idea covering several data communication layers. One of its main components is the Mowgli data channel service (MDCS), which transparently replaces the standard TCP/IP core protocols over the slow wireless link. We discuss how the Mowgli approach, using the MDCS, alleviates the problems encountered with TCP/IP protocols over slow wireless links. The results of our performance tests indicate the merits of the Mowgli approach. The transfer times and the response times become more stable, transfer times for multiple parallel bidirectional transfers are substantially reduced, and response times in interactive work can be kept at a low and predictable level, even when there is other traffic on the wireless link.

Recent developments in mobile communication and personal computer technology have laid a new foundation for mobile computing. Performance of the data communication system as seen by an application program is a fundamental factor when communication infrastructure at the application layer is designed. This paper provides results of performance measurements of data transmission over two different cellular telephone networks, a digital GSM-network and an analogue NMT-network. Since our emphasis is on performance as seen by application programs, we use the standard TCP/IP protocols in the measurements. The performance is measured using three basic operations: establishment of a wireless dial-up connection, exchange of request-reply messages, and bulk data transfer. The external conditions under which the measurements were carried out present a normal office environment when the field strength of the cellular link is good or fairly good.

Modern portable computers and wireless connections over a cellular telephone network have created a new platform for distributed information processing. We present a communication architecture framework which makes it possible to exploit the existing TCP/IP communication architecture but which also takes into account the specific features of wireless links. Our communication architecture is based on the principle of indirect interaction. The mediating interceptor, Mobile-Connection Host, is the bridge between the worlds of wireless and wireline communication. The interceptor also provides enhanced functionality that improves fault-tolerance and performance for applications aware of mobility. Prototypes of the architecture are implemented both for the Unix (Linux) and for the Windows (3.11) platform.

Modern cellular telephone systems extend the usability of portable personal computers enormously. A nomadic user can be given ubiquitous access to remote information stores and computing services. However, the behavior of wireless links creates severe inconveniences within the traditional data communication paradigm. We give an overview of the problems related to wireless mobility. We also present a new software architecture for mastering the problems and discuss a new paradigm for designing mobile distributed applications. The key idea in the architecture is to place a mediator, a distributed intelligent agent, between the mobile node and the wireline network.

Experimnental results are given about the performance of six sorting algorithms in a virtual memory based on the working set principle. With one exception, the algorithms are general internal sorting algorithms and not especially tuned for virtual memory. Algorithms are compared in terms of their time requirements, space requirements, and space-time integrals. The relative performances of the algorithms vary from one measure to the other. Especially in terms of a space-time integral, quicksort turns out to be the best algorithm, also in a working set virtual memory environment.

On the basis of empirical data two topics concerning virtual memory systems are discussed: determining an optimal page size and performance of segmentation as compared to paging. Several production programs have been executed (on a simulator) both in a segmented system and in a paged system with various page sizes; the memory management was based on the working set policy. The memory usage and the fault rates were recorded, and the lifetime functions and space-time integrals were evaluated. The observations are explained using a new model of program behaviour which is a refinement of the phase-transition model. The results show that there is no globally optimal page size. Two characteristic types of programs are observed: the first requires a small page size and a large window size, the second requires a large page size and a small window size. Segmentation and paging are compared with respect to their usage of various resources. In the sense of the space-time integral, segmentation usually outperforms paging; if the mean segment size is large, the difference is remarkable. Several commonly used assumptions about the effects of page size on program behaviour are validated; some of them are found inaccurate or even wrong.

Program restructuring in a segmented virtual memory system is examined on the base of empirical data. The measure of connectivity is based on the decrease of the space-time integral caused by combining the two segments. A heuristic clustering algorithm, based on the use of effectively reduced memory reference strings, is presented. It turns out that performance problems in a segmented system are created by the large number of small segments. Restructuring is able to improve the performance of the system essentially; the optimal performance is reached with values of the memory control parameter which are one magnitude smaller than those applicable to original programs. The clusters formed do not correspond to program phases, but to smaller units of program execution; thus the clustering result supports a hierarchical model of program locality. The restructuring process is reasonably insensitive both to the memory control parameters and to the input data used.

Program behaviour characteristics were examined using data gathered from real program executions. Experiments were performed in a segmented virtual memory with a working set policy; the analyzing costs were kept low using an efficient data reduction method. Empirical results were obtained concerning the influence of the window size on program behaviour characteristics, the accuracy of some average working set size approximations and the sensitivity of program behaviour to the program's input data. These results show that some commonly used assumptions concerning program behaviour are inaccurate. Also there seem to exist “ill-behaving” programs, the behaviour of which does not correspond well with results obtained earlier. The effects of real-time delays during program execution were considered using a new simple method. As an additional experiment, segmenting and paging were compared using various performance statistics; the results seem to favour segmenting.