We consider several well known nonclairvoyant scheduling problems, including the problem of minimizing the average response time, and best-effort firm real-time scheduling. It is known that there are no deterministic online algorithms for these problems with bounded (or even polylogarithmic in the number of jobs) competitive ratios. We show that moderately increasing the speed of the processor used by the nonclairvoyant scheduler effectively gives this scheduler the power of clairvoyance. Furthermore, we show that there exist online algorithms with bounded competitive ratios on all inputs that are not closely correlated with processor speed. 1 Introduction We consider several well known nonclairvoyant scheduling problems, including the problem of minimizing the average response time [13, 15], and besteffort firm real-time scheduling [1, 2, 3, 4, 8, 11, 12, 18]. (We postpone formally defining these problems until the next section.) In nonclairvoyant scheduling some relevant information...

This work describes the design, implementation, and possible applications of Real-Time Linux --- a hard real-time version of the Linux operating system. In this system, a standard time-sharing OS and a real-time executive run on the same computer. Interrupt controller emulation is used to guarantee a low maximum interrupt latency independently of the base system. The use of a one-shot timer makes it possible to achieve a low task release jitter without compromising throughput. Lock-free FIFO buffers are employed for communication between real-time tasks and Linux processes. User-defined schedulers are allowed as are run-time changes in the scheduling policy. The system is in active use for real-time data acquisition, control, and communications. The software is free and can be obtained by FTP at luz.cs.nmt.edu:/pub/rtlinux or via the WWW at http://luz.cs.nmt.edu/~rtlinux. Acknowledgements I would like to express my deep appreciation to my advisor, Victor Yodaiken, for his support, en...

Current research in parallel programming is focused on closing the gap between globally indexed algorithms and the separate address spaces of processors on distributed memory multicomputers. A set of index translation schemes have been implemented as a part of CHAOS runtime support library, so that the library functions can be used for implementing a global index space across a collection of separate local index spaces. These schemes include two software-cached translation schemes aimed at adaptive irregular problems as well as a distributed translation table technique for statically irregular problems. To evaluate and demonstrate the efficiency of the software-cached translation schemes, experiments have been performed with an adaptively irregular loop kernel and a full-fledged 3D DSMC code from NASA Langley on the Intel Paragon and Cray T3D. This paper also discusses and analyzes the operational conditions under which each scheme can produce optimal performance. 1 Introduction Distr...

This paper examines the effects of multithreaded pipelining on the CPI (cycles per instruction) of a RISC processor. The desired CPI in a conventional (single-threaded) RISC processor is one instruction per cycle. However, the CPI is typically more than one because of data hazards, control hazards, and resource hazards in the pipeline.

This dissertation explores the use of statically derived information to improve concurrency control and recovery in objectbase systems. It surveys the relevant background material including existing objectbase systems and conventional concurrency control and recovery. The problem of providing concurrency control and recovery in objectbases supporting nested transactions is analyzed and types of static information which are useful to concurrency control and recovery are determined. Algorithms are proposed to derive the needed static information. The problem of concurrency control in objectbases is decomposed into two simpler problems: intra-transaction concurency control and inter-transaction concurrency control and algorithms are developed for each. A novel concurrency control algorithm combining the algorithms for intra- and inter-transaction concurrency control which specifies serialization orders ` a priori is presented. The algorithm selects appropriate serialization orders based o...

Graph-theoretical tools, algorithm animation, and other re-lated visual-aids have proven very useful in computer- science pedagogy. In this paper the use of the graphical aspects of Petri net theory as a tool to introduce operating system con-cepts is given. It is shown how a minimal amount of Petri net theory cara be applied to problems often discussed in an operating systems course. Examples of models for the con-cept of deadlocks, for the deadlock detection algorithm, and the fork/’oin and parbegin/parend concurrency constructs are shown, In addition, discussion of how Petri net models are utilized to introduce Ada rendezvous solutions of classic problems, such as producer/consumer with a ring-buffer, is given. 1

This qualifying dissertation is intended to review the state-of-the-art of Real-Time Database Systems under a uniprocessor and centralized environments. Due to the heterogeneity of the issues, the large amounts of information, and space limitation, we limit our presentation to the most important issues to the overall design, construction, and advancement of Real-Time Database Systems. Such topics are believed to include Transaction Scheduling, Admission Control, Memory Management, and Disk Scheduling. Furthermore, Transaction Scheduling consists of Concurrency Control Protocols, Conflict Resolution Protocols, and Deadlocks. Out of these issues, the most emphasis is placed on Concurrency Control and Conflict Resolution protocols due to their severe role on the overall systems performance. Other important issues that were not included in our presentation include Fault Tolerance and Failure Recovery, Predictability, and most important of all, Minimizing Transaction Support; i.e., Relaxin...

this paper we shall examine some of the problems facing the World-Wide Web and approaches that may be useful in solving them. In particular, we shall examine problems that relate to the distribution of the Web's information resources. Finally, we shall provide a short evaluation of the Web from the point of view of information providers. It was the intention of the designers of the World-Wide Web (or WWW or W3) to provide access to the information resources of the Internet through easy-to-use software which operated in a consistent manner [BlEtAl92]. As the basis for information retrieval, the designers settled on the hypertext paradigm - a paradigm which supported the use of simple "point-and-click" interfaces.

nt: logic organisation and physical allocation. . Analysis of an existing Operating System: UNIX. Going over and deepening of the user interface: - main user commands: redirection and command piping; - programming by using different command languages: the Bourne shell example. System programming by using the programming language C: in particular, process handling and synchronisation by using the mechanisms of the different versions (Unix BSD 4.3 and Unix System V). . Distributed system and security issues. Communication primitives for UNIX-based distributed systems: the socket mechanism. BIBLIOGRAPHY

processing. The ultimate message from this work should be seen in a larger context. Several languages---such as Ada and Modula-2---are being touted as the standards for the 1990s. These languages severely restrict parallel programming style; this may make saving dusty decks by preprocessing an impossibility. KEY WORDS Parallel programming C CSP Compiler design Distributed systems 0038--0644/92/050371--20$10.00 Received 13 December 1989 1992 by John Wiley & Sons, Ltd. Revised 16 December 1991 372 D. E. STEVENSON ET AL. INTRODUCTION In September 1986, Clemson University purchased a Floating Point Systems (FPS) T-series computer. The T is a hypercube of vector processors: that is, each node is a vector processor. When the T arrived, it had only the occam I language available. In order to make the T usable for numerical work, at least a Fortran compiler was required; certainly for system work, a C compiler was called for. While FPS and Cornell University had already started down th