Mike F. Richardson

Hard Real-Time Scheduling: The Deadline-Monotonic Approach

Abstract The scheduling of processes to meet deadlines is a difficult problem often simplified by placing severe restrictions upon the timing characteristics of individual processes. One restriction often introduced is that processes must have deadline equal to period. This paper investigates schedulability tests for sets of periodic processes whose deadlines are permitted to be less than their period. Such a relaxation enables sporadic processes to be directly incorporated without alteration to the process model. Following an introduction oudining the constraints associated with existing scheduling approaches and associated schedulability tests, the deadline-monotonic approach is introduced. New schedulability tests are derived which vary in computational complexity. The tests are shown to be directly applicable to the scheduling of sporadic processes.

Regulation of GLUT5, GLUT2 and intestinal brush-border fructose absorption by the extracellular signal-regulated kinase, p38 mitogen-activated kinase and phosphatidylinositol 3-kinase intracellular signalling pathways: implications for adaptation to diabetes.

We have investigated the role of the extracellular signal-regulated kinase (ERK), p38 and phosphatidylinositol 3-kinase (PI 3-kinase) pathways in the regulation of intestinal fructose transport. Different combinations of anisomycin, PD98059 and wortmannin had very different effects on fructose transport in perfused isolated loops of rat jejunum. Transport was stimulated maximally by anisomycin (2 microM) and blocked by SB203580 (20 microM), confirming involvement of the p38 pathway. PD98059 (50 microM) alone had little effect on fructose transport. However, it had a dramatic effect on stimulation by anisomycin, diminishing the K(a) 50-fold from 1 microM to 20 nM to show that the ERK pathway restrains the p38 pathway. The K(a) for diabetic jejunum was 30 nM and PD98059 had no effect. Transport in the presence of anisomycin was 3.4-fold that for anisomycin plus PD98059 plus wortmannin. Transport was mediated by both GLUT5 and GLUT2. In general, GLUT2 levels increased up to 4-fold within minutes and with only minimal changes in GLUT5 or SGLT1 levels, demonstrating that GLUT2 trafficks by a rapid trafficking pathway distinct from that of GLUT5 and SGLT1. GLUT2 intrinsic activity was regulated over a 9-fold range. It is concluded that there is extensive cross-talk between the ERK, p38 and PI 3-kinase pathways in their control of brush-border fructose transport by modulation of both the levels and intrinsic activities of GLUT5 and GLUT2. The potential of the intracellular signalling pathways to regulate short-term nutrient transport during the assimilation of a meal and longer-term adaptation to diabetes and hyperglycaemia is discussed.

STRESS: a simulator for hard real-time systems

The STRESS environment is a collection of CASE tools for analysing and simulating the behaviour of hard real‐time safety‐critical applications. It is primarily intended as a means by which various scheduling and resource management algorithms can be evaluated, but can also be used to study the general behaviour of applications and real‐time kernels. This paper describes the structure of the STRESS language and its environment, and gives examples of its use.

Generating feasible cyclic schedules

Abstract The use of a cyclic executive represents the normal implementation scheme for periodic real-time systems, such systems being found in many areas of the control industry. One difficulty with this form of cyclic scheduling is that the schedule itself is difficult to lay out and is fragile in its response to even minor changes to the software it contains. This paper addresses the issues involved in automatically generating, or re-generating, cyclic executives. Four techniques are considered: heuristic search, simulated annealing, stochastic evolution and genetic algorithms, the first two being investigated in detail with prototype tools being implemented and applied to a representative case study. The paper concludes that tool support for this activity is viable. The paper also considers the limitations of static scheduling and considers the use of a more flexible scheme.

Deadline Monotonic Scheduling Theory

Abstract Scheduling theories are now sufficiently mature that a genuine engineering approach to the construction of hard real-time systems is possible. In this paper we discuss the application of Deadline Monotonie Scheduling Theory (DMST). This theory is an extension of the more familiar approach based on rate monotonie priority assignment. The model presented can accommodate periodic and sporadic processes, different levels of criticality, process mteracuon and blocking, precedence constrained processes and multi-deadline processes. It is particularly well integrated with the use of Immediate Priority Ceiling Inheritance for control over process blocking. A basic pseudo-polynomial schedulability test is outlined and then supplemented by the introduction of offsets to control jitter, and period transformation to enable critical (hard) processes to be “protected” during potential transient overloads. These mathematical techniques derived within DMST can help designers experiment with altemauve formulations and prove essenual properties of systems before they arc deployed.

Absolute and relative temporal constraints in hard real-time databases

Application of a conventional database to a hard realtime system requires two inter-related NP-complete problems to be solved: database concurrency control and task scheduling. Starting from temporal constraints on data objects in the database, the techniques described in this paper can be used to reduce the problem to a single NP-complete scheduling problem. Standard scheduling theory can then be applied.

The drtee architecture for distributed hard real-time systems

Abstract The DrTee architecture is aimed at guaranteeing the crucial components of hard real-time distributed systems. A technique of simulated annealing is used to allocate processes to nodes. At each node schedulability tests are employed based on the deadline monotonic approach. At run-time, simple techniques are used: a preemptive priority based scheduler at each node, and a token passing broadcast protocol on the communications bus. Global timing requirements are guaranteed by choosing the priority of each process and the size of the token holding time for each node. The resulting execution environment has the flexibility to cater for sporadic events, mode changes and the dynamic allocate of resources that become available at run-time. These can be used to improve the utility of the services delivered by the real-time system.

A Case Study of the Upper Cotton Valley in the Oak Hill Field of East Texas

This paper summarizes a study of Upper Cotton Valley completions in the northern half of the Oak Hill Field in the East Texas Basin. The study involved the evaluation of full core, sidewall core, special permeability measurements, x-ray diffraction, thin sections, SEM, conventional open hole logs, magnetic resonance imaging tools, micro-resistivity imaging tools, production logs, build-up data, and forty-eight Upper Cotton Valley completions. The overall goal of the project was to improve the completion economics of the Upper Cotton Valley in Oak Hill. This paper details improvements made towards the identification and prediction of productive Upper Cotton Valley sands. An in-depth analysis of the stimulation and completion procedures for the Upper Cotton Valley was also undertaken. That portion of the Upper Cotton Valley study primarily involved the development of a new low sand concentration fracture stimulation technique. These results are presented in a separate paper, SPE 38611. In this paper, a method is described that could be applied to the development of other stacked tight gas sand reservoirs where conventional open hole logs or other conventional evaluation methods have proven to be unreliable. The method resulted in a 50% reduction in average completion costs, with no loss in average initial production rates or reserves per Upper Cotton Valley completion.