A parallel algorithm for preemptive scheduling of uniform machines

In this paper we describe a fast parallel algorithm for preemptive scheduling of n independent jobs on m uniform machines. Each job has a processing requirement, and each machine processes jobs at a different rate. The goal of the scheduling algorithm is to find a schedule which minimizes the time at which the last job is completed. T. Gonzalez and S. Sahni have developed a sequential algorithm which solves this problem in O(n + m log m) time. We develop a parallel version of this algorithm for a Concurrent Read Exclusive Write (CREW) shared memory computer. The algorithm runs in O(log n + log³m) time using n processors.     

Optimal scheduling for two-processor systems

Summary Despite the recognized potential of multiprocessing little is known concerning the general problem of finding efficient algorithms which compute minimallength schedules for given computations and m2 processors. In this paper we formulate a general model of computation structures and exhibit an efficient algorithm for finding optimal nonpreemptive schedules for these structures on two-processor systems. We prove that the algorithm gives optimal solutions and discuss its application to preemptive scheduling disciplines.     

A 2-approximation NC algorithm for connected vertex cover and tree cover

The connected vertex cover problem is a variant of the vertex cover problem, in which a vertex cover is additional required to induce a connected subgraph in a given connected graph. The problem is known to be NP-hard and to be at least as hard to approximate as the vertex cover problem is. While several 2-approximation NC algorithms are known for vertex cover, whether unweighted or weighted, no parallel algorithm with guaranteed approximation is known for connected vertex cover. Moreover, converting the existing sequential 2-approximation algorithms for connected vertex cover to parallel ones results in RNC algorithms of rather high complexity at best.In this paper we present a 2-approximation NC (and RNC) algorithm for connected vertex cover (and tree cover). The NC algorithm runs in O(log²n) time using O(Δ²(m+n)/logn) processors on an EREW-PRAM, while the RNC algorithm runs in O(logn) expected time using O(m+n) processors on a CRCW-PRAM, when a given graph has n vertices and m edges with maximum vertex degree of Δ.     

Online scheduling of moldable parallel tasks

In this paper, we study an online scheduling problem with moldable parallel tasks on m processors. Each moldable task can be processed simultaneously on any number of processors of a parallel computer, and the processing time of a moldable task depends on the number of processors allotted to it. Tasks arrive one by one. Upon arrival of each task, the scheduler has to determine both the number of processors and the starting time for the task. Moreover, these decisions cannot be changed in the future. The objective is to attain a schedule such that the longest completion time over all tasks, i.e., the makespan, is minimized. First, we provide a general framework to show that any \(\rho \)-bounded algorithm for scheduling of rigid parallel tasks (the number of processors for a task is fixed a prior) can be extended to yield an algorithm for scheduling of moldable tasks with a competitive ratio of \(4\rho \) if the ratio \(\rho \) is known beforehand. As a consequence, we achieve the first constant competitive ratio, 26.65, for the moldable parallel tasks scheduling problem. Next, we provide an improved algorithm with a competitive ratio of at most 16.74.     

Scheduling Independent Multiprocessor Tasks

We study the problem of scheduling a set of n independent multiprocessor tasks with prespecified processor allocations on a fixed number of processors. We propose a linear time algorithm that finds a schedule of minimum makespan in the preemptive model, and a linear time approximation algorithm that finds a schedule of makespan within a factor of (1+\eps) of optimal in the non-preemptive model. We extend our results by obtaining a polynomial time approximation scheme for the parallel processors variant of the multiprocessor task model.     

Deadline-based scheduling of periodic task systems on multiprocessors

We consider the problem of scheduling periodic task systems on multiprocessors and present a deadline-based scheduling algorithm for solving this problem. We show that our algorithm successfully schedules on m processors any periodic task system with utilization at most m²/(2m−1).     

Efficient scheduling of tasks without full use of processor resources

The nonpreemptive scheduling of a partially ordered set of tasks on a machine with m processors of different speeds is studied. Polynomial time algorithms are presented which benefit from selective non-use of slow processors. The performance of these heuristics is asymptotic to √m times worse than optimal, whereas list schedules are unboundedly worse than optimal for any fixed value of m. The techniques of analyzing these schedules are used to obtain a bound on a large class of preemptive schedules.     

Scheduling parallel jobs to minimize the makespan

We consider the NP-hard problem of scheduling parallel jobs with release dates on identical parallel machines to minimize the makespan. A parallel job requires simultaneously a prespecified, job-dependent number of machines when being processed. We prove that the makespan of any nonpreemptive list-schedule is within a factor of 2 of the optimal preemptive makespan. This gives the best-known approximation algorithms for both the preemptive and the nonpreemptive variant of the problem. We also show that no list-scheduling algorithm can achieve a better performance guarantee than 2 for the nonpreemptive problem, no matter which priority list is chosen. List-scheduling also works in the online setting where jobs arrive over time and the length of a job becomes known only when it completes; it therefore yields a deterministic online algorithm with competitive ratio 2 as well. In addition, we consider a different online model in which jobs arrive one by one and need to be scheduled before the next job becomes known. We show that no list-scheduling algorithm has a constant competitive ratio. Still, we present the first online algorithm for scheduling parallel jobs with a constant competitive ratio in this context. We also prove a new information-theoretic lower bound of 2.25 for the competitive ratio of any deterministic online algorithm for this model. Moreover, we show that 6/5 is a lower bound for the competitive ratio of any deterministic online algorithm of the preemptive version of the model jobs arriving over time.     

Preemptive scheduling with release times and deadlines

We consider the problem of deciding if there is a feasible preemptive schedule for a set of n independent tasks with release times and deadlines on m identical processors. The general problem is known to be solvable in O(n ³) time. In this paper, we study special cases for which faster algorithms exist. We introduce the notion of monotone schedules and study their properties. These properties are then exploited to devise fast algorithms for two special cases—the nested task systems and the non-overlapping task systems. We give an O(n log mn) time algorithm and an O(n log n+mn) time algorithm for nested task systems and non-overlapping task systems, respectively. Our algorithms generate at most O(n) and O(mn) preemptions for nested task systems and nonoverlapping task systems, respectively.Research supported in part by the ONR grant N00014-87-K-0833.     

Optimal preemptive semi-online scheduling on two uniform processors

We investigate a preemptive semi-online scheduling problem. Jobs with sizes within a certain range [1,r] (r?1) arrive one by one to be scheduled on two uniform parallel processors with speed 1 and s?1, respectively. The objective is to minimize makespan. We characterize the optimal competitive ratio as a function of both s and r by devising a deterministic on-line scheduling algorithm along with a matching lower bound, which also holds for randomized algorithms.     

Parallel algorithms for arrangements

We give the first efficient parallel algorithms for solving the arrangement problem. We give a deterministic algorithm for the CREW PRAM which runs in nearly optimal bounds ofO (logn log^* n) time andn ²/logn processors. We generalize this to obtain anO (logn log^* n)-time algorithm usingn ^d /logn processors for solving the problem ind dimensions. We also give a randomized algorithm for the EREW PRAM that constructs an arrangement ofn lines on-line, in which each insertion is done in optimalO (logn) time usingn/logn processors. Our algorithms develop new parallel data structures and new methods for traversing an arrangement.This work was supported by the National Science Foundation, under Grants CCR-8657562 and CCR-8858799, NSF/DARPA under Grant CCR-8907960, and Digital Equipment Corporation. A preliminary version of this paper appeared at the Second Annual ACM Symposium on Parallel Algorithms and Architectures [3].     

Preemptive scheduling of independent jobs on identical parallel machines subject to migration delays

We present hardness and approximation results for the problem of preemptive scheduling of n independent jobs on m identical parallel machines subject to a migration delay d with the objective to minimize the makespan. We give a sharp threshold on the value of d for which the complexity of the problem changes from polynomial time solvable to NP-hard. Next, we give initial results supporting a conjecture that there always exists an optimal schedule with at most m − 1 job migrations. Finally, we provide a O(n) time (1 + 1/log₂ n)-approximation algorithm for m = 2.     

Scheduling independent jobs on uniform parallel machines to minimize tardiness criteria

The problem of scheduling N jobs on M uniform parallel machines is studied. The objective is to minimize the mean tardiness or the weighted sum of tardiness with weights based on jobs, on periods or both. For the mean tardiness criteria in the preemptive case, this problem is NP-hard but good solutions can be calculated with a transportation problem algorithm. In the nonpreemptive case the problem is therefore NP-hard, except for the cases with equal job processing times or with job due dates equal to job processing times. No dominant heuristic is known in the general nonpreemptive case. The author has developed a heuristic to solve the nonpreemptive scheduling problem with unrelated job processing times. Initially, the algorithm calculates a basic solution. Next, it considers the interchanges of job subsets to equal processing time sum interchanging resources (i.e. a machine for a given period). This paper models the scheduling problem. It presents the heuristic and its result quality, solving 576 problems for 18 problem sizes. An application of school timetable scheduling illustrates the use of this heuristic.     

Parallel integer sorting and simulation amongst CRCW models

In this paper a general technique for reducing processors in simulation without any increase in time is described. This results in an O(√logn) time algorithm for simulating one step of PRIORITY on TOLERANT with processor-time product of O(n log logn); the same as that for simulating PRIORITY on ARBITRARY. This is used to obtain anO(logn/log logn + √logn (log logm ? log logn)) time algorithm for sortingn integers from the set {0,...,m ? 1},m ≧n, with a processor-time product ofO(n log logm log logn) on a TOLERANT CRCW PRAM. New upper and lower bounds for ordered chaining problem on an allocated COMMON CRCW model are also obtained. The algorithm for ordered chaining takesO(logn/log logn) time on an allocated PRAM of sizen. It is shown that this result is best possible (upto a constant multiplicative factor) by obtaining a lower bound of Ω(r logn/(logr + log logn)) for finding the first (leftmost one) live processor on an allocated-COMMON PRAM of sizen ofr-slow virtual processors (one processor simulatesr processors of allocated PRAM). As a result, for ordered chaining problem, “processor-time product” has to be at least Ω(n logn/log logn) for any poly-logarithmic time algorithm. Algorithm for ordered-chaining problem results in anO(logN/log logN) time algorithm for (stable) sorting ofn integers from the set {0,...,m ? 1} withn-processors on a COMMON CRCW PRAM; hereN = max(n, m). In particular if,m =n ^O(1), then sorting takes Θ(logn/log logn) time on both TOLERANT and COMMON CRCW PRAMs. Processor-time product for TOLERANT isO(n(log logn)²). Algorithm for COMMON usesn processors.     

Assigning real-time tasks on heterogeneous multiprocessors with two unrelated types of processors

Consider the problem of partitioned scheduling of an implicit-deadline sporadic task set on heterogeneous multiprocessors to meet all deadlines. Each processor is either of type-1 or type-2. We present a new algorithm, FF-3C, for this problem. FF-3C offers low time-complexity and provably good performance. Specifically, FF-3C offers (i) a time-complexity of O(n?max(m,logn)+m?logm), where n is the number of tasks and m is the number of processors and (ii) the guarantee that if a task set can be scheduled by an optimal partitioned-scheduling algorithm to meet all deadlines then FF-3C meets all deadlines as well if given processors at most $\frac{1}{1-\alpha}$ times as fast (referred to as speed competitive ratio) and tasks are scheduled using EDF; where α is a property of the task set. The parameter α is in the range (0,0.5] and for each task, it holds that its utilization is no greater than α or greater than 1?α on each processor type. Thus, the speed competitive ratio of FF-3C can never exceed 2. We also present several extensions to FF-3C; these offer the same performance guarantee and time-complexity but with improved average-case performance. Via simulations, we compare the performance of our new algorithms and two state-of-the-art algorithms (and variations of the latter). We evaluate algorithms based on (i) running time and (ii) the necessary multiplication factor, i.e., the amount of extra speed of processors that the algorithm needs, for a given task set, so as to succeed, compared to an optimal task assignment algorithm. Overall, we observed that our new algorithms perform significantly better than the state-of-the-art. We also observed that our algorithms perform much better in practice, i.e., the necessary multiplication factor of the algorithms is much smaller than their speed competitive ratio. Finally, we also present a clustered version of the new algorithm.     

Scheduling Independent Multiprocessor Tasks

Abstract. We study the problem of scheduling a set of n independent multiprocessor tasks with prespecified processor allocations on a fixed number of processors. We propose a linear time algorithm that finds a schedule of minimum makespan in the preemptive model, and a linear time approximation algorithm that finds a schedule of makespan within a factor of (1+\eps) of optimal in the non-preemptive model. We extend our results by obtaining a polynomial time approximation scheme for the parallel processors variant of the multiprocessor task model.     

Minimization of earliness, tardiness and due date penalties on uniform parallel machines with identical jobs

We consider a problem of scheduling n identical nonpreemptive jobs with a common due date on m uniform parallel machines. The objective is to determine an optimal value of the due date and an optimal allocation of jobs onto machines so as to minimize a total cost function, which is the function of earliness, tardiness and due date values. For the problem under study, we establish a set of properties of an optimal solution and suggest a two-phase algorithm to tackle the problem. First, we limit the number of due dates one needs to consider in pursuit of optimality. Next, we provide a polynomial-time algorithm to build an optimal schedule for a fixed due date. The key result is an O(m² log m) algorithm that solves the main problem to optimality.Scope and purpose: To extend the existing research on cost minimization with earliness, tardiness and due date penalties to the case of uniform parallel machines.     

Nonpreemptive Coordination Mechanisms for Identical Machines

We focus on the problem of scheduling n weighted selfish tasks on m identical parallel machines and we study the performance of nonpreemptive coordination mechanisms. A nonpreemptive coordination mechanism consists of m local scheduling policies that decide the processing order of the tasks on each machine without delays or interruptions. We discuss the existence of Nash equilibria for this setting and show that it is not a guaranteed property of all nonpreemptive coordination mechanisms. Next, we focus on the wider class of randomized Nash equilibria and prove lower bounds on the price of anarchy. Our lower bounds are presented in comparison to the currently best known coordination mechanism (which uses delays) and lead to the conclusion that preemption or delays are required in order to improve on the currently best known solution.     

Task graph pre-scheduling, using Nash equilibrium in game theory

Prescheduling algorithms are targeted at restructuring of task graphs for optimal scheduling. Task graph scheduling is a NP-complete problem. This article offers a prescheduling algorithm for tasks to be executed on the networks of homogeneous processors. The proposed algorithm merges tasks to minimize their earliest start time while reducing the overall completion time. To this end, considering each task as a player attempting to reduce its earliest time as much as possible, we have applied the idea of Nash equilibrium in game theory to determine the most appropriate merging. Also, considering each level of a task graph as a player, seeking for distinct parallel processors to execute each of its independent tasks in parallel with the others, the idea of Nash equilibrium in game theory can be applied to determine the appropriate number of processors in a way that the overall idle time of the processors is minimized and the throughput is maximized. The communication delay will be explicitly considered in the comparisons. Our experiments with a number of known benchmarks task graphs and also two well-known problems of linear algebra, LU decomposition and Gauss–Jordan elimination, demonstrate the distinguished scheduling results provided by applying our algorithm. In our study, we consider ten scheduling algorithms: min–min, chaining, A ^?, genetic algorithms, simulated annealing, tabu search, HLFET, ISH, DSH with task duplication, and our proposed algorithm (PSGT).     

An NC algorithm for finding a minimum weighted completion time schedule on series parallel graphs

We present a parallel algorithm for solving the minimum weighted completion time scheduling problem for transitive series parallel graphs. The algorithm takesO(log² n) time withO(n ³) processors on a CREW PRAM, wheren is the number of vertices of the input graph. This is the first NC algorithm for solving the problem.Research supported in part by NSF Grants CCR-9011214 and CCR-9205982.