The need for supportingvariety of hard and soft real-time, as well as best effort applications in a multimedia computing environment requires an operating system framework that: (1) enables different schedulers to be employed for different application classes, and (2) provides protection between the various classes of applications. We argue that these objectives can be achieved by hierarchical partitioning of CPU bandwidth, in which an operating system partitions the CPU bandwidth among various application classes, and each application class, in turn, partitions its allocation (potentially using a different scheduling algorithm) among its sub-classes or applications. We present Start-time Fair Queuing (SFQ) algorithm, which enables such hierarchical partitioning. We have implemented a hierarchical scheduler in Solaris 2.4. We describe our implementation, and demonstrate its suitability for multimedia operating systems. 1

this paper, we describe an adaptive QOS mapping scheme where the QOS parameters of applications are mapped into resource requirements dynamically, and the resources necessary for the applications are reserved based on the measured resource utilization. The approach combines the merits of the both approaches described in the previous paragraph. Real-Time Mach provides resource reservation facilities that are suitable for supporting the adaptive QOS mapping scheme. The first resource reservation mechanism is the processor reservation mechanism that reserves CPU resources for respective applications. The second resource reservation mechanism is the memory reservation that reserves free pages for respective applications. We describe these resource reservation mechanisms, and show the effectiveness of the mechanisms. The remainder of this paper is structured as follows. In Section 2, we describe the adaptive QOS mapping scheme. Section 3 presents the processor reservation mechanism on Real-time Mach, and our memory reservation mechanism is described in Section 4. In Section 5, we discuss some issues about our proposed mechanisms. Section 5 summarizes the paper.

A QOS control is one of the most important factors in continuous media applications. The QOS levels of the applications should be maintained even if several continuous media applications are running concurrently. Also, the QOS levels should be changed according to a number of applications and their respective importances when a system is overloaded. This paper presents our experience with a video-ondemand system that supports a dynamic QOS control scheme on Real-Time Mach, and shows the necessity of a real-time resource management for building such applications. 1 Introduction Multimedia computing has emerged in the last few years as a major area of computer science. This work is motivated by a wide range of potential applications made feasible by integrating continuous media such as audio and video in current computing environment. Now, many modern workstations are equipped with microphones and speakers, and varieties of software packages are available in areas such as music composi...

Increases in processor speeds and the availability of audio and video devices for personal computers have encouraged the development of interactive multimedia applications for teleconferencing and digital audio/video presentation among others. These applications have stringent timing constraints, and traditional operating systems are not well suited to satisfying such constraints. On the other hand, hard real-time systems that can meet these constraints are typically static and inflexible. This dissertation

This paper proposes a virtual memory management system suitable for interactive continuous media applications. Interactive continuous media applications usually require a large amount of memory for storing their code, data, and stack segments. In traditional operating systems, demand paging makes it possible to execute such large applications by storing most pages in secondary disks. However, continuous media applications should avoid page faults for ensuring timing constraints of continuous media since it takes a long time to swap pages between physical memory and secondary storages. Thus, it is difficult to satisfy timing constraints of continuous media. Therefore, some operating systems provide memory wiring primitives that enable applications to wire pages in physical memory by specifying the range of virtual address spaces explicitly. On the other hand, our virtual memory management system enables continuous media applications to reserve physical memory for allocating pages as soo...

Data Structures ...........................................................162 6.3. The Graph Extraction Algorithm ..............................................169 7. The RTM Virtual Machine ....................................... 180 7.1. Initial Concerns ..........................................................................180 7.2. The Virtual Machine Abstraction ..............................................182 7.3. The System Level........................................................................184 7.4. The Distributed-System Level....................................................186 7.5. The Local Machine Component Level .......................................189 7.6. The Virtual-Machine Interface Functions .................................193 vii 8. The Prototype Implementation............................... 196 8.1. Mach Features ............................................................................196 8.2. The Mach/486 RTM Implementation .................

Multimedia computing has emerged in the last few years as a major area of computer science. However, multimedia programming is very hard since programmers need to take into account many complex facilities such as real-time processing, media synchronization, and dynamic QOS control. Since usual programmers take a long time to understand these facilities, multimedia toolkits that can hide such complex facilities from programmers should be provided for making multimedia programming easier. The toolkit should also provide mechanisms for implementing large continuous media applications. Especially, programmers may expect to reuse existing applications for building their applications quickly. If continuous media applications can be constructed by composing a small number of big modules, programmers can create continuous media applications very quickly. In this paper, we describe a continuous media toolkit that has been developing at Multimedia and Mobile Computing Group at Japan Advanced Ins...

: Multimedia and advanced real-time systems integrate a variety of applications and media types in one environment, many of which have timing properties which are not supported by current window systems running in multitasking environments. This paper discusses several key aspects to designing window systems which support continuous media and other real-time applications. These design considerations were used to construct ARTIFACT, a window system designed to investigate issues in building real-time window systems. Several of the behaviors of a prototype implementation of ARTIFACT are described, along with preliminary scheduling models and avenues for further research. Keywords: real-time systems, operating system support, window systems, multimedia support, scheduling theory 1. Introduction Multimedia systems integrate a variety of media types into a unified display environment. Some media types, such as text and static graphics, have been in use for quite some time. Other types, su...

In this paper, we describe an adaptive QOS mapping scheme where the QOS parameters of applications are mapped into resource requirements dynamically, and the resources for the applications are reserved based on the measured resource utilization. Real-Time Mach provides resource reservation facilities that are suitable for supporting the adaptive QOS mapping scheme, and we show the effectiveness of the resource reservation facilities when the adaptive QOS mapping scheme is adopted. 1 Introduction Resource reservation such as processor reservation and memory reservation is a promising approach for supporting timing critical applications such as continuous media applications[1, 4, 7, 11, 12]. However, it is difficult to predict how much resource is consumed for respective applications before executing the applications. In realtime communities, the schedulability analysis can be used to map QOS values of continuous media applications to their resource utilization. The approach is realisti...

NPS is a real-time network engine provided by RealTime Mach for real-time applications and distributed multimedia systems. NPS provides a framework for implementing real-time network protocols which are able to bound end-to-end delay and throughput. Real-Time Mach also provides a framework to select the suitable implementations of a network protocol implemented in different network engines. 1 Introduction Real-time communication is becoming a more important part of distributed systems requiring rigid timing constraints such as distributed multimedia systems. These applications require for throughput, end-to-end delay and jitter to be bounded. The network systems which ensure such requirements are realized by the combinations of new infrastructures such as ATM[10], new protocols[1, 2] and new network engines, where network engines provide the execution environment for network protocols. Current trends of researches on network engines, however, focus on high performance[6, 12] and flex...