Coupled tasks scheduling was originally introduced for modelling complex radar devices. It is still used for controlling such devices and applied in similar applications. This paper considers a problem of coupled tasks scheduling on one processor, under the assumptions that all processing times are equal to 1, the gap has a constant exact length and the precedence constraints are strict. Although it is proven that the problem stated above is NP-hard in the strong sense if the precedence constraints have a form of a general graph, it is possible to solve some of its relaxed versions in polynomial time. This paper contains a solution for the problem of coupled tasks scheduling with an assumption that the precedence constraints graph has a form of chains and it presents an algorithm that can solve the problem with such assumption in time O(n log n).

The coupled tasks scheduling problem was originally introduced for modeling complex radar devices. It is still used for controlling such devices and applied in similar applications. This paper considers a problem of coupled tasks scheduling on a single processor, under the assumptions that all processing times are equal to 1, the gap has exact integer length L and the precedence constraints are strict. We prove that the general problem, when L is part of the input and the precedence constraints graph is a general graph, is NP-hard in the strong sense. We also show that the special case when L = 2 and the precedence constraints graph is an in-tree or an out-tree, can be solved in O(n) time.

An important concern in the attempt of understanding the functional code of eukaryotic genes is to elucidate the control structures for regulating gene activation and suppression. One objective in the attempt to understand mechanisms of gene regulation is the elucidation of the regulatory network structure. A preliminary step of a detailed network analysis is identifying the transcription factor binding sites of a regulatory network. Known as cis-regulatory module (CRM), it is understood as part of the genome that comprises a set of short length binding sites. Gene regulatory systems are known to be quite stable during evolution, as compared to relatively frequent replication processes of genes and mutations of the coding sequences. This conservation property of regulatory code can advantageously be used for identifying cis-regulatory modules of potentially co-regulated genes. As the degree of similarity is expected to depend on the phylogenetic distance of homologs or orthologs, we favor an approach that is based on a comparison paradigm. The paper introduces a novel concept for measuring the similarity ofcis-regulatory modules which can then be used in an algorithm for comparing regulatory regions. The proposed algorithm searches for pairs of similar modules, and a prototype implementation is applied to human and mouse liver sequences. The results are compared to that of random sequences, and it is shown that a clear decision about co-regulation is possible at this level.

This paper addresses the problem of resource allocation for distributed real-time periodic tasks, operating in environments that undergo unpredictable changes and that defy the specification of meaningful worst-case execution times. These tasks are supplied by input data originating from various environmental workload sources. Rather than using worst-case execution times (WCETs) to describe the CPU usage of the tasks, we assume here that execution profiles are given to describe the running time of the tasks in terms of the size of the input data of each workload source. The objective of resource allocation is to produce an initial allocation that is robust against fluctuations in the environmental parameters. We try to maximize the input size (workload) that can be handled by the system, and hence to delay possible (costly) reallocations as long as possible. We present an approximation algorithm based on first-fit and binary search that we call FFBS. As we show here, the first-fit algorithm produces solutions that are often close to optimal. In particular, we show analytically that FFBS is guaranteed to produce a solution that is at least 41% of optimal, asymptotically, under certain reasonable restrictions on the running times of tasks in the system. Moreover, we show that if at most 12% of the system utilization is consumed by input independent tasks (e.g., constant time tasks), then FFBS is guaranteed to produce a solution that is at least 33% of optimal, asymptotically. Moreover, we present simulations to compare FFBS approximation algorithm with a set of standard (local search) heuristics such as hill-climbing, simulated annealing, and random search. The results suggest that FFBS, in combination with other local improvement strategies, may be a reasonable approach for resource allocation in dynamic real-time systems.

This paper discusses the problem of controlling highly dynamic technical facilities such as mechatronic systems. To enforce the required behavior, sensors informing about the status of the facility are periodically read. The sensor signals are processed in end-to-end computations by computational activities, and, in case the measured data deviate from the technical specification, result in new settings of actuators. We consider the case that the changes of operating conditions of the facility underlie continuous mode changes. It is assumed that the variety of operation modes mechatronic systems may work in are restricted in the sense that a higher requirement in one system component may exclude higher requirements in other components. We expect that such dependencies are specified in the technical specification of the facility, and that from them the domain of realistic modes can be derived. The objective of this work is defining heuristic algorithms that intend to finding minimum sets of feasible allocations for a fixed number of processors, and comparing their usefulness experimentally.

In this paper we are concerned with algorithms producing schedules for processes which consist of a finite set of tasks and a precedence relation between them. For executing tasks a system of m identical processors is available. We are interested in algorithms producing non preemptive schedules which are nearly time-optimal. For some well-known algorithms the lengths of their schedules are compared with the lengths of time-minimal schedules. A new algorithm which produces considerably better schedules is presented and discussed.

This work deals with the problem of controlling highly dynamic mechatronic systems. Such systems may work in several different operation modes, or even underlie continuous mode changes. While concepts are available that deal with discrete operation modes, the continuous case is unsolved. In This work, some ideas of how to tackle the scheduling problem with continuous mode changes are discussed. It is assumed that in mechatronic systems not all modes are realistic. A higher requirement in one system component may exclude higher requirements in other components. We expect that such dependencies are specified in the technical specification of the mechatronic system, and from them the domain of realistic modes can be derived. We discuss two optimization problems: (1) given a set of hosts, find a minimum number of allocations that feasibly cover the whole domain of realistic modes. (2) In the design phase one would like to minimize the number of processors for which such a set of allocations exists.

During the last years considerable effort has been put into developing scheduling and load balancing techniques and models for the processing of tasks in parallel and distributed systems. In this paper we extend an existing scheduling model called the divisible task model to make it applicable to a broader class of practical applications. To achieve this we present two new modifications to give the model a greater flexibility and to increase its performance. The goal of this work is to reduce the transfer volume over the communication links and getting an equally balanced load on the nodes of the system. As test-system we use a cluster-based image database to check the achieved performance.

view of our impossibility result it might also be of interest to study the impact of the amount of locality information in the static case, and in the dynamic case, in which locality refers also to the dimension of time, that is, the amount of local information regarding the occurrence of the messages. 26 follow only one path between them (as is the case in certain situations, like in certain ATM routing problems), then the resulting problem is NP-complete, even when the network is bottleneck-free, and the processors have global information about the initial connguration. Theorem 15 Given a network G = (V; E), given capacities c(e) for every edge e 2 E, and given integers a i;j 0, for i; j 2 V , it is NP-Complete to determine whether there exists a feasible schedule in which, for every i; j 2 V , a i;j messages are sent from i to j, all following the same path between i and j, such that each meets its given deadline and such that no more than c(e) messages are delivered along any edge e in any unit of time. Proof: We use a reduction from the Partition problem (see 5]): Given a set of integers fa i ji = 1; ; 2ng, is it possible to nd a subset I f1; ; 2ng such that P i2I a i = P i= 2I a i ? Given an instance of this problem, we deene a network, in which processor a is connected to processor b by two parallel lines e 1 and e 2 , and b is connected to 2n processors p 1 through p 2n. It is required to send a i messages from a to p i , for i = 1; ; 2n. The capacity of each link is 1 2 P 1i2n a i , and the deadline for each message is 1. It is possible to deliver all messages within one time unit if and only if some messages can be sent on link e 1 and the rest of them on link e 2 at time 0, and then at time 1 each message is sent by b to its destination, thus meeting its deadline. Due to the capacities of the links e 1 and e 2 , this can happen if and only if there is a subset I f1; ; 2ng …

Considering a decision support system as a tool where executive's judgment can be included along with the mathematical tool kit of the management scientist, this paper shows the need to include problem management as an integral component of the decision support system for scheduling problems. A methodology based on the resolution of conflicts among various constraints in scheduling problems is proposed to implement the problem management system in a decision support system for these problems. The paper concludes with some guidelines to create a workable framework for providing effective decision support to solve scheduling problems and the identification of some fruitful directions for future research.

One of the most important problems arising in multiprocessor systems is scheduling of tasks on a set of parallel processors. Recently, new models of task processing have been formulated in which certain tasks can require more than one processor at a time. This model is especially justified in some applications of multi-microprocessor systems. In this paper, we extend the above model to cover the case of scheduling in the presence of additional scarce resources. First a subcase of the problem is considered in which preemptable tasks need simultaneously one or two processors and one additional resource in the amount of one unit. For this case a low order polynomial-time algorithm is presented. Then the general case is solved via a linear programming approach.

In this paper we are concerned with the subproblem of bin packing, where the set of possible weights of elements is finite. In [5] it was mentioned that this problem could be solved by an exhaustive search procedure in polynomial time, but the degree of the polynomial is high and increases as the cardinality of the set of weights increases. However, we will show that a more careful analysis of the problem leads to a linear time algorithm. The impact of this result on task scheduling is discussed.