Cloud computing is now widely used in all major industries. This transformative technology allows users to access the resources they need through a pay-as-you-go model, providing an efficient and convenient service. Although cloud computing can provide users with flexible resource scheduling and payment models, improper use may lead to cost overruns. In addition to makespan, the total cost of cloud services is also a key user requirement. Therefore, in this paper, we propose a heuristic algorithm aimed at solving the optimization problem of workflow scheduling with deadline constraints in cloud environments in order to minimize the total cost. The PACM algorithm assigns the corresponding deadline to each task in the workflow in upward probabilistic order, and achieves a more reasonable deadline assignment for the tasks based on the Critical Path Factor and Dependency Amplification Factor. Finally, the tasks are then assigned to cloud services to meet their sub-deadline requirements. Experiments are conducted using well-known scientific workflows for performance evaluation, and the results show that the algorithm outperforms previous heuristic algorithms, proving the effectiveness of the algorithm.

Multi-view clustering (MVC) exploits the information captured from diverse views to partition data into different groups and attracts much attention recently. Despite significant progress, most MVC methods fuse multi-view information via one-stage fusion while neglecting the merits of multi-stage fusion which causes insufficient in utilizing rich information within data and therefore degrades the clustering performance. To this end, designing a functional framework that can fully exploit multi-view information becomes a key challenge in multi-view clustering research. In this paper, we propose a novel multi-stage fusion method, which elegantly unifies the late and early fusion into one unified framework, to capture sufficient information underlying the multi-view data and to effectively reduce the effect of low-quality views. Specifically, we construct a low dimensional latent representation from multi-view data by learning proper correlation among multi-view data in the early fusion stage. The late fusion establishes a new optimal combinational data partition from base partitions constructed by spectral clustering, which suppresses the influence of low-quality basic partitions. Then we couple the low dimensional latent representation with the learned combinational data partition to share the same cluster structure by$k$-means and maximization alignment. As a result, we collaboratively learn an accurate and robust partition representation for the following clustering task. Besides, the late fusion and early fusion are jointly learned to achieve mutual collaboration for better performance. Finally, an alternating optimization algorithm is designed to solve the resultant optimization problem. Extensive experiments conducted on eight datasets show the superiority of our method in terms of effectiveness and efficiency.

High reliability and low latency are conflicting when tasks are scheduled. Scheduling of parallel applications with data dependencies in heterogeneous systems is an NP-complete problem. Using replication to improve system reliability can lead to increased application execution time. From the perspective of increasing the reliability of real-time heterogeneous system considering communication overhead and the timing requirements, this paper proposed a fault-tolerant scheduling algorithm based on local maximum reliability replication strategy (FTSA-BLMR). Our algorithm first sets the maximum number of replications for each task. Then it continuously replicates the task with the highest system reliability for the current task set to obtain a new task set. The tasks in the new task set will be scheduled and the scheduling results will be recorded. Finally, the scheduling result will be selected as the final scheduling sequence, which has maximum system reliability and meets the deadline. The experimental results indicate that our algorithm can improve the system reliability by 40% compared with DB-FTSA when the deadline constraint is strict.

In a heterogeneous multi-core real-time system, once a permanent error occurs, the task cannot be successfully completed before the deadline, which may cause catastrophic consequences. Therefore, the reliability of the real-time system is critical. In this paper, we consider real-time tasks in a heterogeneous system with previous constraints, and explore how to improve the reliability of the system. We propose a new scheduling algorithm-PFTSA, which uses active replication to back up as many tasks on different processors as possible before the deadline, minimizing communication overhead, ensuring that the maximum number of permanent errors can be accommodated before the deadline and providing maximum system reliability. The experimental results show that the reliability of our proposed scheduling algorithm is higher than the existing related algorithms.

Due to the increasing demand for computing power in many industries, heterogeneous multi-core processors are needed to solve the problem. In order to make full use of multi-core computing resources, an effective scheduling strategy for heterogeneous multi-core processor tasks is required. Directed acyclic graph (DAG) is usually used to represent data dependencies between tasks. Each task needs to be executed in the order of its data dependencies. Research under this model has made great progress. In this article, we study and improve the DAG-based task model, taking into account the fact that not only one-way data transmission is possible between tasks, but also two-way data exchange. Based on this model, we propose two scheduling strategies, overall cutting scheduling (OCS) and greedy selection scheduling (GSS). As far as we know, there is currently no work considering the existence of a special task model of two-way transmission between tasks, nor has it considered task scheduling in two-way transmission. In order to evaluate and demonstrate its feasibility and practicability, we proposed a reference method and supplemented with large-scale system experiments. These experiments show that the scheduling efficiency of the proposed method is greatly improved.

With the gradual complexity and high parallelization of computing tasks, people have higher and higher requirements for the reliability and energy-saving performance of embedded systems. In this paper, we focus on the scheduling of parallel application in heterogeneous systems. Based on task replication and DVFS techniques, we propose two algorithms ERO and ORO that could reduce energy consumption while meeting the task reliability requirements. Comparative experiments with EFSRG and HRRM shown that ERO and ORO algorithms provide more energy savings and lower schedule lengths for the given reliability requirement.

For a set of periodic real-time tasks running on a multi-processor system, some tasks need access to shared resources, while the remaining tasks do not. This article aims to solve the problem of priority inversion caused by simultaneous access to shared resources by tasks in a multi-processor real-time system. We propose a task allocation model and partition scheduling algorithm based on the MSRP protocol, which is called SASR-MSRP. Firstly, the algorithm divides the task set into two categories based on whether the task accesses shared resources or not. Secondly, calculate the system utilization rate U of the task that accesses the shared resource and determine the execution priority of the task according to its non-increasing order and assign it to the corresponding processor. Finally, we use the EDF scheduling algorithm to sequentially allocate the remaining independent tasks to the idle time period of the application processor. This algorithm not only reduces the problem of priority inversion, but also improves the overall scheduling efficiency of the system.

When the traditional heterogeneous multi-core scheduling algorithm performs tasks with high resource density, a large amount of idle time often occurs on the processor core. Therefore, based on the environment of heterogeneous multi-core processors, this paper studies the static heuristic table scheduling algorithm, and proposes an optimization approach for the problem of single priority assignment and too simple task assignment. We design optimization in the static heuristic scheduling algorithm list generation phase and task allocation phase, and propose a hybrid task allocation method with three strategies to improve the standby time utilization of processor core. Then, DVFS technology is used to optimize the scheduling results, so that the task can run with lower energy consumption without increasing makespan. Finally, the new algorithm is compared with three traditional scheduling algorithms through design experiments, and it is proved that the new algorithm has better performance when executing more tasks.