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.     

Optimal online-list batch scheduling

We consider the online-list batch scheduling problem. Jobs arrive one by one and have to be assigned upon arrival to a scheduled batch such that the makespan is minimized. Each batch can accommodate up to B jobs. We give a complete classification of the tractability of this online problem.     

Dynamic scheduling of a batch of parallel task jobs on heterogeneous clusters

This paper addresses the problem of minimizing the scheduling length (make-span) of a batch of jobs with different arrival times. A job is described by a direct acyclic graph (DAG) of parallel tasks. The paper proposes a dynamic scheduling method that adapts the schedule when new jobs are submitted and that may change the processors assigned to a job during its execution. The scheduling method is divided into a scheduling strategy and a scheduling algorithm. We also propose an adaptation of the Heterogeneous Earliest-Finish-Time (HEFT) algorithm, called here P-HEFT, to handle parallel tasks in heterogeneous clusters with good efficiency without compromising the makespan. The results of a comparison of this algorithm with another DAG scheduler using a simulation of several machine configurations and job types shows that P-HEFT gives a shorter makespan for a single DAG but scores worse for multiple DAGs. Finally, the results of the dynamic scheduling of a batch of jobs using the proposed scheduler method showed significant improvements for more heavily loaded machines when compared to the alternative resource reservation approach.     

Online scheduling of two job types on a set of multipurpose machines with unit processing times

We study a problem of scheduling a set of n jobs with unit processing times on a set of m multipurpose machines in which the objective is to minimize the makespan. It is assumed that there are two different job types, where each job type can be processed on a unique subset of machines. We provide an optimal offline algorithm to solve the problem in constant time and an online algorithm with a competitive ratio that equals the lower bound. We show that the worst competitive ratio is obtained for an inclusive job-machine structure in which the first job type can be processed on any of the m machines while the second job type can be processed only on a subset of m/2 machines. Moreover, we show that our online algorithm is 1-competitive if the machines are not flexible, i.e., each machine can process only a single job type.     

Preemptive online scheduling with rejection of unit jobs on two uniformly related machines

We consider preemptive online and semi-online scheduling of unit jobs on two uniformly related machines. Jobs are presented one by one to an algorithm, and each job has a rejection penalty associated with it. A new job can either be rejected, in which case the algorithm pays its rejection penalty, or it can be scheduled preemptively on the machines, in which case it may increase the maximum completion time of any machine in the schedule, also known as the makespan of the constructed schedule. The objective is to minimize the sum of the makespan of the schedule of all accepted jobs and the total penalty of all rejected jobs. We study two versions of the problem. The first one is the online problem where the jobs arrive unsorted, and the second variant is the semi-online case, where the jobs arrive sorted by a non-increasing order of penalties. We also show that the variant where the jobs arrive sorted by a non-decreasing order of penalties is equivalent to the unsorted one. We design optimal online algorithms for both cases. These algorithms have smaller competitive ratios than the optimal competitive ratio for the more general problem with arbitrary processing times (except for the case of identical machines), but larger competitive ratios than the optimal competitive ratio for preemptive scheduling of unit jobs without rejection.     

Scheduling with multiple servers

In this paper, we consider the problem of scheduling a set of jobs on a set of identical parallel machines. Before the processing of a job can start, a setup is required which has to be performed by a given set of servers. We consider the complexity of such problems for the minimization of the makespan. For the problem with equal processing times and equal setup times we give a polynomial algorithm. For the problem with unit setup times, m machines and m − 1 servers, we give a pseudopolynomial algorithm. However, the problem with fixed number of machines and servers in the case of minimizing maximum lateness is proven to be unary NP-hard. In addition, recent algorithms for some parallel machine scheduling problems with constant precessing times are generalized to the corresponding server problems for the case of constant setup times. Moreover, we perform a worst case analysis of two list scheduling algorithms for makespan minimization.     

Approximation Algorithms for Multiprocessor Scheduling under Uncertainty

Motivated by applications in grid computing and project management, we study multiprocessor scheduling in scenarios where there is uncertainty in the successful execution of jobs when assigned to processors. We consider the problem of multiprocessor scheduling under uncertainty, in which we are given n unit-time jobs and m machines, a directed acyclic graph C giving the dependencies among the jobs, and for every job j and machine i, the probability p _ij of the successful completion of job j when scheduled on machine i in any given particular step. The goal of the problem is to find a schedule that minimizes the expected makespan, that is, the expected time at which all of the jobs are completed.     

Scheduling on two identical machines with a speed-up resource

We address a variant of scheduling problem on two identical machines, where we are given an additional speed-up resource. If a job uses the resource, its processing time may decrease. However, at any time the resource can only be used by at most one job. The objective is to minimize the makespan. For the offline version, we present an FPTAS. For the online version where jobs arrive over list, we propose an online algorithm with competitive ratio of 1.781, and show a lower bound of 1.686 for any online algorithm.     

Non-clairvoyant Scheduling Games

In a scheduling game, each player owns a job and chooses a machine to execute it. While the social cost is the maximal load over all machines (makespan), the cost (disutility) of each player is the completion time of its own job. In the game, players may follow selfish strategies to optimize their cost and therefore their behaviors do not necessarily lead the game to an equilibrium. Even in the case there is an equilibrium, its makespan might be much larger than the social optimum, and this inefficiency is measured by the price of anarchy—the worst ratio between the makespan of an equilibrium and the optimum. Coordination mechanisms aim to reduce the price of anarchy by designing scheduling policies that specify how jobs assigned to a same machine are to be scheduled. Typically these policies define the schedule according to the processing times as announced by the jobs. One could wonder if there are policies that do not require this knowledge, and still provide a good price of anarchy. This would make the processing times be private information and avoid the problem of truthfulness. In this paper we study these so-called non-clairvoyant policies. In particular, we study the RANDOM policy that schedules the jobs in a random order without preemption, and the EQUI policy that schedules the jobs in parallel using time-multiplexing, assigning each job an equal fraction of CPU time.     

Minimising makespan for two batch-processing machines with non-identical job sizes in job shop

In this article, the job shop scheduling problem with two batch-processing machines is considered. The machines have limited capacity and the jobs have non-identical job sizes. The jobs are processed in batches and the total size of each batch cannot exceed the machine capacity. The processing times of a job on the two machines are proportional. We show the problem of minimising makespan is NP-hard in the strong sense. Then we provide an approximation algorithm with worst-case ratio no more than 4, and the running time of the algorithm is O(n?log?n). Finally, the performance of the proposed algorithm is tested by different levels of instances. Computational results demonstrate the effectiveness of the algorithm for all the instances.     

Online scheduling on semi-related machines

We study the problem of minimizing the makespan on related machines in the following setting: jobs arrive over time, and the machines may become available or unavailable. In either case, advance warning is given, i.e., the next point of time where a job is released or a machine changes state is revealed a little time ahead. We present an optimal online algorithm for this problem.     

Scheduling with conflicts: online and offline algorithms

We consider the following problem of scheduling with conflicts (swc): Find a minimum makespan schedule on identical machines where conflicting jobs cannot be scheduled concurrently. We study the problem when conflicts between jobs are modeled by general graphs. Our first main positive result is an exact algorithm for two machines and job sizes in {1,2}. For jobs sizes in {1,2,3}, we can obtain a -approximation, which improves on the -approximation that was previously known for this case. Our main negative result is that for jobs sizes in {1,2,3,4}, the problem is APX-hard. Our second contribution is the initiation of the study of an online model for swc, where we present the first results in this model. Specifically, we prove a lower bound of on the competitive ratio of any deterministic online algorithm for m machines and unit jobs, and an upper bound of 2 when the algorithm is not restricted computationally. For three machines we can show that an efficient greedy algorithm achieves this bound. For two machines we present a more complex algorithm that achieves a competitive ratio of when the number of jobs is known in advance to the algorithm.     

An exact algorithm for parallel machine scheduling with conflicts

We consider an extension of classic parallel machine scheduling where a set of jobs is scheduled on identical parallel machines and an undirected conflict graph is part of the input. Each node in the graph represents a job, and an edge implies that its two jobs are conflicting, meaning that they cannot be scheduled on the same machine. The goal is to find an assignment of the jobs to the machines such that the maximum completion time (makespan) is minimized. We present an exact algorithm based on branch and price that combines methods from bin packing, scheduling, and graph coloring, with appropriate modifications. The algorithm has a good computational performance even for parallel machine scheduling without conflicting jobs.     

Online scheduling with rejection and withdrawal

We study an online scheduling problem with rejection, in which some rearrangement of the solution is allowed. This problem is called scheduling with rejection and withdrawal. Each arriving job has a processing time and a rejection cost associated with it, and it needs to be either assigned to a machine or rejected upon arrival. At termination, it is possible to choose at most a fixed number of scheduled jobs and withdraw them (i.e., decide to reject them). We study the minimization version, where the goal is to minimize the sum of the makespan and the total rejection cost (which corresponds to the penalty), and the maximization problem, where the goal is to maximize the sum of the minimum load and the total rejection cost (which corresponds to profit). We study environments of machines, which are the case of m identical machines and the case of two uniformly related machines, and show a strong relation between these problems and the related classic online scheduling problems which they generalize, in contrast to standard scheduling with rejection, which typically makes the scheduling problems harder.     

A multi-level scheduler for batch jobs on grids

This paper proposes a two-level scheduler for dynamically scheduling a continuous stream of sequential and multi-threaded batch jobs on grids, made up of interconnected clusters of heterogeneous single-processor and/or symmetric multiprocessor machines. The scheduler aims to schedule arriving jobs respecting their computational and deadline requirements, and optimizing the hardware and software resource usage. At the top of the hierarchy a lightweight meta-scheduler (MS) classifies incoming jobs according to their requirements, and schedules them among the underlying resources balancing the workload. At cluster level a Flexible Backfilling algorithm carries out the job machine associations by exploiting dynamic information about the environment. Scheduling decisions at both levels are based on job priorities computed by using different sets of heuristics. The different proposals have been compared through simulations. Performance figures show the feasibility of our approach.     

Scheduling jobs with equal processing times subject to machine eligibility constraints

We consider the problem of nonpreemptively scheduling a set of n jobs with equal processing times on m parallel machines so as to minimize the makespan. Each job has a prespecified set of machines on which it can be processed, called its eligible set. We consider the most general case of machine eligibility constraints as well as special cases of nested and inclusive eligible sets. Both online and offline models are considered. For offline problems we develop optimal algorithms that run in polynomial time, while for online problems we focus on the development of optimal algorithms of a new and more elaborate structure as well as approximation algorithms with good competitive ratios.     

Scheduling jobs on identical machines with agreement graph

We consider the following problem of scheduling with agreements: a set of jobs must be scheduled non-preemptively on identical machines subject to constraints that only some specific jobs can be scheduled concurrently on different machines. These constraints are represented by an agreement graph and the aim is to minimize the makespan. This problem is NP-hard. We study the complexity of the problem for two machines and arbitrary bipartite agreement graphs, in particular we prove the NP-hardness of the open problem proposed in the literature which is the case of two machines with processing times at most 3. We propose list algorithms with empirical results for the problem in the general case.     

带等级约束的多重工件在线(半在线)排序问题

论文提出了带等级约束的多重工件排序问题,每个客户提交多个加工时间和等级相同的工件。目标是寻找一个调度方案,使得机器的最大完工时间最小。当客户的信息未知时,论文设计了一个竞争比为5/3的在线算法。当所有工件的加工时间总和已知时,论文设计了一个竞争比为3/2的半在线算法。这些结论对经典带等级约束的两台平行机排序问题进行了推广。     

Online hierarchical scheduling: An approach using mathematical programming

This paper considers an online hierarchical scheduling problem on parallel identical machines. We are given a set of m machines and a sequence of jobs. Each machine has a different hierarchy, and each job also has a hierarchy associated with it. A job can be assigned to a machine only if its hierarchy is no less than that of the machine. The objective is to minimize the makespan, i.e., the maximum load of all machines. Two models are studied in the paper. For the fractional model, we present an improved algorithm and lower bounds. Both the algorithm and the lower bound are based on solutions of mathematical programming. For any given m, our algorithm is optimal by numerical calculation. For the integral model, we present both a general algorithm for any m, and an improved algorithm with better competitive ratios of 2.333 and 2.610 for m=4 and 5, respectively.