Multiprocessor Scheduling Research Articles

Offline routing and spectrum allocation algorithms for elastic optical networks

Elastic optical networks (EONs) have emerged as a promising solution for future high speed networks because of their ability to efficiently manage network resources and provide better spectrum utilization to cope with the recent rapid change in traffic behavior and the tremendous growth in bandwidth demand. The routing and spectrum allocation (RSA) problem is one of the key challenges for the effective design and control of EONs. Recently, the offline RSA problem has been mapped to a multiprocessor scheduling problem and solved using a well-known list scheduling heuristic. In this paper, we address an RSA multiprocessor scheduling formulation with the objective of minimizing the total amount of spectrum needed to serve the traffic demand in chain and mesh networks. The quality of the solution using list-scheduling algorithms is very sensitive to the ordering of the tasks in the list. We propose four list-ordering algorithms (the longest then widest compact algorithm, area compact algorithm, longest then left edge compact algorithm, and area then left edge compact algorithm) to enhance the performance of an existing list-scheduling algorithm called the compact scheduling algorithm. Our proposed algorithms are based on the left edge algorithm and the combination of the problem dimensions (i.e., the bandwidth and links). We evaluate the performance and the efficiency of our proposed algorithms across a range of demand distributions for two different network topologies (i.e., chain and National Science Foundation networks).Experimental results show that our algorithms outperform existing algorithms and provide close to optimal solutions, which are within 1–2% of the lower bound.

From RUN to QPS: new trends for optimal real-time multiprocessor scheduling

Until recently, there has been a common belief that optimal multiprocessor real-time scheduling algorithms necessarily incur a high number of task preemptions and migrations. New scheduling algorithms have shown that this is not the case. In this paper, we explain why two of these algorithms, reduction to uniprocessor (RUN) and quasi-partition scheduling (QPS), achieve optimality with only a few preemptions and migrations, exhibiting their similarities and differences. We also compare these two algorithms via simulations, highlighting the consequences of their differences in their performance. By putting RUN and QPS side-by-side, we bring about their fundamental properties and help in the understanding of the multiprocessor real-time scheduling problem.

EDZL Scheduling and Schedulability Analysis for Performance Asymmetric Multiprocessors

Heterogeneous multiprocessor architectures allow embedded real-time systems to better match computing resources to each application's needs and dynamic workload requirements, thereby providing many opportunities for improved performance with reduced power consumption. Unfortunately, guaranteeing real-time requirements on heterogeneous multiprocessors remains a critical problem due to the lack of appropriate scheduling algorithms and analysis methods. In this paper, we consider EDZL (Earliest Deadline First until Zero-Laxity) for performance asymmetric multiprocessor scheduling. EDZL has been shown to outperform other scheduling policies such as global EDF on identical multiprocessors. We show that EDZL is still effective on performance asymmetric multi-processors, and present an efficient EDZL schedulability test. Experimental results show that EDZL scheduling is able to schedule up to 20% more task sets than global EDF and that our new EDZL schedulability test can accept up to 30% more schedulable task sets than a presently exiting one.

When is the immune inspired B-Cell algorithm superior to the (1+1) evolutionary algorithm?

There exist many experimental investigations of artificial immune systems (AIS), and it has been shown that the AIS is useful and efficient for many real-world optimisation problems. However, we know little about that whether the AIS can outperform the traditional evolutionary algorithms on some optimisation problems in theory. This work rigorously proved that a simple AIS called the B-cell algorithm (BCA) with somatic contiguous hypermutations can efficiently optimise two instances of the multiprocessor scheduling problem in expected polynomial runtime, whereas the local search algorithms and the (1+1) evolutionary algorithm ((1+1) EA) using only one individual in the search space and with standard bit mutation are highly inefficient. This work is helpful for gaining insight into the idea there exists no algorithm which is efficient for all specific problems.

From RUN to QPS: new trends for optimal real-time multiprocessor scheduling

From RUN to QPS: new trends for optimal real-time multiprocessor scheduling

Planning computations in a multiprocessor system with unfixed parameters

The problem of constructing a feasible preemptive multiprocessor schedule is considered for a case where directive intervals are assigned, processors can have arbitrary performance, and the amount of tasks depends linearly on the volume of additional resources allocated for them. In cases where a feasible schedule is not found with an allocated volume of additional resources, the problem of optimally correcting the directive intervals is considered. The solution is based on an analysis of the necessary and sufficient conditions of a feasible schedule’s existence.

Thread-level priority assignment in global multiprocessor scheduling for DAG tasks

The advent of multi- and many-core processors offers enormous performance potential for parallel tasks that exhibit sufficient intra-task thread-level parallelism. With a growth of novel parallel programming models (e.g., OpenMP, MapReduce), scheduling parallel tasks in the real-time context has received an increasing attention in the recent past. While most studies focused on schedulability analysis under some well-known scheduling algorithms designed for sequential tasks, little work has been introduced to design new scheduling policies that accommodate the features of parallel tasks, such as their multi-threaded structure. Motivated by this, we refine real-time scheduling algorithm categories according to the basic unit of scheduling and propose a new priority assignment method for global task-wide thread-level fixed-priority scheduling of parallel task systems. Our evaluation results show that a finer-grained, thread-level fixed-priority assignment, when properly assigned, significantly improves schedulability, compared to a coarser-grained, task-level assignment.

LBBA: An efficient online benefit-aware multiprocessor scheduling for QoS via online choice of approximation algorithms

Maximizing the benefit gained by soft real-time jobs in many applications and embedded systems is highly needed to provide an acceptable QoS (Quality of Service). This paper considers a benefit model for on-line preemptive multiprocessor scheduling. The goal is to maximize the total benefit gained by the jobs that meet their deadlines. This method prioritizes the jobs using their benefit density functions and schedules them in a real-time basis. We propose an online choice of two approximation algorithms in order to partition the jobs among identical processors at the time of their arrival without using any statistics. Our analysis and experiments show that we are able to maximize the gained benefit and decrease the computational complexity (compared to existing algorithms) while minimizing makespan (response time, also referred to as cost), with fewer missed deadlines and more balanced usage of processors. Our solution is applicable to a wide variety of soft real-time applications and embedded systems such as, but not limited to multimedia applications, medical monitoring systems or those with higher utilization such as bursty hosting servers.11This research is done by Behnaz Sanati and supervised by Professor Albert Cheng.

Enhanced Preemptive Global Utility Accrual Real Time Scheduling Algorithms in Multicore Environment

This paper proposed an efficient real time scheduling algorithm using global scheduling paradigm running in multicore environment known as Global Preemptive Utility Accrual Scheduling (GPUAS) algorithm. The existing TUF/UA multiprocessor scheduling algorithms known as Greedy-Global Utility Accrual (G-GUA) and Non Greedy-Global Utility Accrual (NG-GUA) algorithms is seen to overlook the efficiency on its task scheduling algorithm. These algorithms have adapted the task migration attribute considering the load balancing problem in multi core platform. The existing PUAS uniprocessor scheduling algorithm is mapped into the multicore scheduling environment that consists of the global scheduling schemes considering the migration attribute of the executed tasks. The main principal of global scheduling is that it allows the executed tasks to migrate from one processor to the other processors whenever a scheduling event occurs in the system. The proposed GPUAS algorithm inherits the characteristics of PUAS in uniprocessor where it can preempt the highest PUD task at any event that occurs in the system. In this research, the proposed GPUAS algorithm enhanced the existing NG-GUA and G-GUA algorithms. The developed simulator has derived the set of parameter, events and performance metrics according to a detailed analysis of the base model. The proposed GPUAS algorithm achieved the highest accrued utility for the entire load range. The proposed GPUAS algorithm is more efficient than the existing algorithms, producing the highest accrued utility ratio and less abortion ratio making it more suitable and efficient for real time application domain.

A WEAKLY HARD SCHEDULING APPROACH OF PARTITIONED SCHEDULING ON MULTIPROCESSOR SYSTEMS

Real-time systems or tasks can be classified into three categories, based on the “seriousness” of deadline misses – hard, soft and weakly hard real-time tasks. The consequences of a deadline miss of a hard real-time task can be prohibitively expensive because all the tasks must meet their deadlines whereas soft real-time tasks tolerate “some” deadline misses. Meanwhile, in a weakly hard real-time task, the distribution of its met and missed deadlines is stated and specified precisely. As real-time application systems increasingly come to be implemented upon multiprocessor environments, thus, this study applies multiprocessor scheduling approach for verification of weakly hard real-time tasks and to guaranteeing the timing requirements of the tasks. In fact, within the multiprocessor, the task allocation problem seem even harder than in uniprocessor case; thus, in order to cater that problem, the sufficient and efficient scheduling algorithm supported by accurate schedulability analysis technique is present to provide weakly hard real-time guarantees. In this paper, a weakly hard scheduling approach has been proposed and schedulability analysis of proposed approach consists of the partitioned multiprocessor scheduling techniques with solutions for the bin-packing problem, called R-BOUND-MP-NFRNS (R-BOUND-MP with next-fit-ring noscaling) combining with the exact analysis, named hyperperiod analysis and deadline models; weakly hard constraints and µ-pattern under static priority scheduling. Then, Matlab simulation tool is used in order to validate the result of analysis. From the evaluation results, it can be proven that the proposed approach outperforms the existing approaches in terms of satisfaction of the tasks deadlines.

Multiprocessor Scheduling And Performance Evaluation Using Elitist Non Dominated Sorting Genetic Algorithm for Independent Task

Multiprocessor Scheduling And Performance Evaluation Using Elitist Non Dominated Sorting Genetic Algorithm for Independent Task

Multiprocessor Scheduling And Performance Evaluation Using Elitist Non Dominated Sorting Genetic Algorithm for Independent Task

Multiprocessor Scheduling And Performance Evaluation Using Elitist Non Dominated Sorting Genetic Algorithm for Independent Task

On the compatibility of exact schedulability tests for global fixed priority pre-emptive scheduling with Audsley’s optimal priority assignment algorithm

Audsley's optimal priority assignment (OPA) algorithm can be applied to multiprocessor scheduling provided that three conditions hold with respect to the schedulability tests used. In this short paper, we prove that no exact test for global fixed priority pre-emptive scheduling of sporadic tasks can be compatible with Audsley's algorithm, and hence the OPA algorithm cannot be used to obtain an optimal priority assignment for such systems.

Bin Packing Algorithms with Applications to Passenger Bus Loading and Multiprocessor Scheduling Problems

Bin Packing Algorithms with Applications to Passenger Bus Loading and Multiprocessor Scheduling Problems

Task Scheduling and Idle-Time Balancing in Homogeneous Multi Processors: A Comparison between GA and SA

scheduling problem has a special significance in multiprocessors due to efficient use of the processor and also spending less time. Tasks should be assigned to processors in such a way to minimizing makespan. In this paper, we use genetic algorithm and simulated annealing to solve task scheduling problem on multi homogenous processors with minimizing completion time. In addition we introduce another fitness function as processors idle-time balancing which should be less than a predetermined value. These algorithms are used to determine suitable priorities that lead to a sub- optimal solution. And finally to compare the performance of these algorithms, we design 9 test problem based on two fitness function.

On the complexity of constructing multiprocessor little-preemptive schedules

We present a full and correct proof of the fact that the problem of constructing an optimal schedule for the open shop problem with at most m − 3 preemptions for an m-processor system is NP-hard. We also show that the proof of this result given by E. Shchepin and N. Vakhania in Ann. Oper. Res. 159, 183–213 (2008) is incorrect.

An unfair semi-greedy real-time multiprocessor scheduling algorithm

Most real-time multiprocessor scheduling algorithms for achieving optimal processor utilization, adhere to the fairness rule. Accordingly, tasks are executed in proportion to their utilizations at each time quantum or at the end of each time slice in a fluid schedule model. Obeying the fairness rule results in a large number of scheduling overheads, which affect the practicality of the algorithm. This paper presents a new algorithm for scheduling independent real-time tasks on multiprocessors, which produces very few scheduling overheads while maintaining high schedulability. The algorithm is designed by totally relaxing the fairness rule and adopting a new semi-greedy criterion instead. Simulations have shown promising results, i.e. the scheduling overheads generated by the proposed algorithm are significantly fewer than those generated by state-of-the-art algorithms. Although the proposed algorithm sometimes misses a few deadlines, these are sufficiently few to be tolerated in view of the considerable reduction achieved in the scheduling overheads.

English

English

End-to-End Communication Delay Analysis in Industrial Wireless Networks

WirelessHART is a new standard specifically designed for real-time and reliable communication between sensor and actuator devices for industrial process monitoring and control applications. End-to-end communication delay analysis for WirelessHART networks is required to determine the schedulability of real-time data flows from sensors to actuators for the purpose of acceptance test or workload adjustment in response to network dynamics. In this paper, we consider a network model based on WirelessHART, and map the scheduling of real-time periodic data flows in the network to real-time multiprocessor scheduling. We then exploit the response time analysis for multiprocessor scheduling and propose a novel method for the delay analysis that establishes an upper bound of the end-to-end communication delay of each real-time flow in the network. Simulation studies based on both random topologies and real network topologies of a <inline-formula><tex-math>$74$</tex-math></inline-formula> -node physical wireless sensor network testbed demonstrate that our analysis provides safe and reasonably tight upper bounds of the end-to-end delays of real-time flows, and hence enables effective schedulability tests for WirelessHART networks.

Static Task Scheduling Algorithm with Minimum Distance for Multiprocessor System (STMD)

Task scheduling in the multiprocessor environment is considered to be a NP-complete problem. It is also called a multiprocessor scheduling. Here, a parallel program is divided into a number of subtasks and it is mapped on the processors for concurrent execution. The objective of task scheduling is to reduce scheduling length of a given application program. Considering that task scheduling is represented by Directed Acyclic Grape (DAG). There are two types of task scheduling algorithms: static task scheduling and dynamic task scheduling algorithms. In this paper, we propose a new task scheduling algorithm based on minimum distance and an entry task duplicated on the processors. The minimum distance computes between an entry task and its successors. It includes, minimum distance of parent task, execution time and communication time Here , also considering a priority list (PL) attribute that contains tasks information of a given application program whose distance is minimum from an entry task. Finally, we will do comparative study of the proposed algorithm with Bounded Number of Processors (BNP) class of scheduling algorithms. It has done based on the following matrices: Scheduling Length, Speedup, Efficiency, Load Balancing and Normalized Scheduling Length (NSL). It showed proposed algorithm gives better result than BNP class of scheduling algorithms.