Derek McAuley

The design and implementation of an operating system to support distributed multimedia applications

Support for multimedia applications by general purpose computing platforms has been the subject of considerable research. Much of this work is based on an evolutionary strategy in which small changes to existing systems are made. The approach adopted is to start ab initio with no backward compatibility constraints. This leads to a novel structure for an operating system. The structure aims to decouple applications from one another and to provide multiplexing of all resources, not just the CPU, at a low level. The motivation for this structure, a design based on the structure, and its implementation on a number of hardware platforms is described.

The desk area network

A novel architecture for use within an end computing system is described. This attempts to extend the concepts used in modern high speed networks into computer system design. A multimedia workstation is being built based on this concept to evaluate the approach.

Energy is just another resource: energy accounting and energy pricing in the Nemesis OS

In this paper, we argue that, with an appropriate operating system structure, energy in mobile computers can be treated and managed as just another resource. In particular, we investigate how energy management could be added to the Nemesis OS which provides detailed and accurate resource accounting capabilities in order to provide Quality of Service (QoS) guarantees for all resources to applications. We argue that, with such an operating system, accounting of energy to individual processes can be achieved. Furthermore, we investigate how an economic model, proposed for congestion avoidance in computer network, and recently applied to CPU resource management, can be used as a dynamic, decentralised energy management system, forming a collaborative environment between operating system and applications.

Differential QoS and pricing in networks: where flow control meets game theory

This paper looks at ways of providing quality of service (QoS) to users, based on a simple pricing scheme. It is primarily aimed at elastic traffic, and it is users rather than the network who define the flow control schemes. A framework for assessing schemes and algorithms via a distributed game is presented.

Automics: souvenir generating photoware for theme parks

Automics is a photo-souvenir service which utilises mobile devices to support the capture, sharing and annotation of digital images amongst groups of visitors to theme parks. The prototype service mixes individual and group photo-capture with existing in-park, on-ride photo services, to allow users to create printed photo-stories. Herein we discuss initial fieldwork in theme parks that grounded the design of Automics, our development of the service prototype, and its real-world evaluation with theme park visitors. We relate our findings on user experience of the service to a literature on mobile photoware, finding implications for the design of souvenir services.

ATM everywhere

The use and deployment of asynchronous transfer mode (ATM) in a target of operating environments and administrative domains are discussed. The operating environments include wide area ATM, regional area ATM, local area ATM, desk area ATM, and local wireless ATM. The challenges of interdomain ATM are reviewed.<<ETX>>

Integrated optical 2 /spl times/ 2 switch for wavelength multiplexed interconnects

A highly compact integrated optical switch is proposed and demonstrated for broadband optical switching applications. Routing of 8 /spl times/ 10 Gb/s data channels is demonstrated using a low-cost 1250-Mb/s control scheme. The advantages of lossless operation, broad optical bandwidth, and nanosecond switching times are leveraged. Multichannel wavelength is exploited for reduced latency, enhanced capacity, and functionality, while retaining compatibility with existing off-the-shelf electronics and transceiver technology. The requirements for optical header processing, wavelength translation, and optical buffering are avoided. Low-penalty multiwavelength transmission is demonstrated for a highly compact sub-mm/sup 2/ amplifying 2 /spl times/ 2 switch. Pattern dependent gain and amplified spontaneous emission are minimized to facilitate 0.0-0.4 dB penalty. Mitigation techniques compatible with the architecture are deployed to reduce the penalty under adverse operating conditions. Control schemes are proposed and demonstrated to facilitate 8 /spl times/ 10 Gb/s optically switched networking.

Breath control of amusement rides

Emerging robotic technologies are enabling the control of individual seats on rollercoasters and other thrill rides. We explore the potential of breathing as an effective and engaging way of driving this. Observations and interviews from trials of an enhanced bucking bronco ride show that breath-control is fun, challenging and intelligible, and reveal riders-x tactics as they battled the machine. We conclude that breath control is feasible and appropriate for controlling rides, unpack its important characteristics, and consider how it might be built into future ride systems. We argue that the combination of voluntary and involuntary factors in breathing is especially appealing for controlling rides as it balances game-like elements of skill and learning against the thrill of surrendering control to the machine.

A case for virtual channel processors

Modern desktop and server computer systems use multiple processors: general purpose CPU(s), graphic processor (GPU), network processors (NP) on Network Interface Cards (NICs), RAID controllers, and signal processors on sound cards and modems. Some of these processors traditionally have been special purpose processors but there is a trend towards replacing some of these with embedded general purpose processors. At the same time main CPUs become more powerful; desktop CPUs start featuring Simultaneous Multi-Threading (SMT); and Symmetric Multi-Processing (SMP) systems are widely used in server systems. However, the structure of operating systems has not really changed to reflect these trends --- different types of processors evolve at different timescales (largely driven by market forces) requiring significant changes to operating systems kernels to reflect the appropriate tradeoffs.In this position paper we propose to re-vitalise the old idea of channel processors by encapsulating operating system I/O subsystems in Virtual Channel Processors (VCPs). VCPs perform I/O operations on behalf of an OS. They provide similar development, performance, and fault isolation as dedicated (embedded) I/O processors do while offering the flexibility to split functionality between the main processor(s) and dedicated processors without affecting the rest of the OS. If part of a VCP is executed on the main processor, we propose to make use of virtual machine technology and SMT/SMP features to isolate its performance from that of the rest of the system and to protect the system from faults within the VCP.

Experiences of building an ATM switch for the local area

The Fairisle project was concerned with ATM in the local area. An earlier paper [9] described the preliminary work and plans for the project. Here we present the experiences we have had with the Fairisle network, describing how implementation has changed over the life of the project, the lessons learned, and some conclusions about the work so far.