Search

Recently, wide-area distributed computing environments have become popular owing to their huge resource capability. In a wide-area distributed computing environment, joint scheduling of tasks and data is the main strategy to improve system performance. However, the geographically distributed diverse resources exhibit high variations, making it challenging to design efficient joint scheduling of tasks and data. To accurately adapt to the dynamic variations of geographically distributed diverse resources and achieve a high system performance, this study proposes a hypergraph-partitioning-based online joint scheduling method. The proposed method constructs a hypergraph of geographically distributed tasks, data, and diverse resources to clearly describe the correlation among the three elements and quantitatively reflect the time cost of different process in the environment. The hypergraph is dynamically updated according to the generated scheduling scheme and the collected information to reflect the dynamic variations of resource states. Then, a hypergraph partition optimization mechanism is proposed to generate efficient joint scheduling schemes, thus reducing the overall completion time in the system. The experimental results indicate that compared with the state-of-the-art joint scheduling methods, the proposed method reduces the overall completion time by up to 25.67% and significantly reduces the task waiting time, although it makes a concession in the data migration time. [ABSTRACT FROM AUTHOR]

Tasks have high requirements for response delay and security in intelligent manufacturing. Industrial data have the characteristics of high privacy. However, cloud services are difficult to implement for low latency-sensitive applications and privacy data tasks. Therefore, the offloading technology in edge computing can offload the computing tasks of terminal devices to the edge of the network, which can effectively reduce the delay and match the needs of intelligent manufacturing. Unreasonable task scheduling cannot meet the needs of real-time scheduling between edge servers and cloud servers. In this paper, we establish a joint low-delay optimization model of task scheduling and dynamic replacement-release caching (DRRC) mechanism, which couples a privacy selection strategy for tasks to protect privacy. Tasks are scheduled to different location by the privacy of sensitive data, which can improve the security of data and meet the calculation request of different tasks. DRRC mechanism caches tasks according to the size of the task and replaces it with the weight of the task data, and adds automatic release mechanism. To solve the task scheduling strategy, we design the improved genetic-differential evolution algorithm. Extensive simulations reveal that the proposed algorithm has a better performance in minimizing latency compared with other scheduling algorithms. At the same time, the caching mechanism has a better hit rate. [ABSTRACT FROM AUTHOR]

The central cloud facilities based on virtual machines offer many benefits to reduce the scheduling costs and improve service availability and accessibility. The approach of cloud computing is practical due to the combination of security features and online services. In the tasks transfer, the source and target domains have differing feature spaces. This challenge becomes more complicated in network traffic, which leads to data transfer delay, and some critical tasks could not deliver at the right time. This paper proposes an efficient optimization method for task scheduling based on a hybrid multi-verse optimizer with a genetic algorithm called MVO-GA. The proposed MVO-GA is proposed to enhance the performance of tasks transfer via the cloud network based on cloud resources' workload. It is necessary to provide adequate transfer decisions to reschedule the transfer tasks based on the gathered tasks' efficiency weight in the cloud. The proposed method (MVO-GA) works on multiple properties of cloud resources: speed, capacity, task size, number of tasks, number of virtual machines, and throughput. The proposed method successfully optimizes the task scheduling of a large number of tasks (i.e., 1000–2000). The proposed MVO-GA got promising results in optimizing the large cloud tasks' transfer time, which reflects its effectiveness. The proposed method is evaluated based on using the simulation environment of the cloud using MATLAB distrusted system. [ABSTRACT FROM AUTHOR]

The cloud runtime environment is dynamic; therefore, allocating tasks to computing resources might include various scenarios. Metaheuristic algorithms are usually used to choose appropriate scheduling scenarios; however, they suffer from premature convergence, trapping in local optima, and imbalance between the exploration and exploitation of search space. The multi-verse optimizer (MVO) algorithm also suffers from similar problems. In this research, both Gabor filter and opposition-based learning methods are applied in the MVO algorithm to present the new algorithm GOMVO. The multi-tracker optimization (MTO) is applied in the GOMVO to present the new MTO-GOMVO hybrid algorithm. Then the scheduling framework MTOA-GOMVO@DSF is presented that applies the MTO-GOMVO metaheuristic algorithms in cloud computing scheduling. In the sequel, at first, the GOMVO algorithm is benchmarked applying CEC2017 benchmark functions and compared with several baseline algorithms in terms of mean error. Second, MTOA-GOMVO is also evaluated against related baseline algorithms in terms of mean error. Finally, MTOA-GOMVO is also applied in cloud computing to schedule independent tasks to virtual machines to improve average execution time, response time, throughput, and SLA violations. Simulation results applying NASA-iPSC real dataset showed that MTOA-GOMVO outweighs the baseline metaheuristic algorithms and performs well in scheduling cloud computing tasks. [ABSTRACT FROM AUTHOR]

Recent advancements of virtualization technologies for parallel processing involve scheduling containerized tasks in a workflow. Since a container can include multiple tasks, it can be reused or shared among applications. If every task in a workflow uses its dedicated container without sharing among any tasks, each container image must be downloaded for each task. As a result, many computational resources are required to process and the communication latency related to container image downloading can become a bottleneck for the makespan. In task scheduling algorithms for workflows, this characteristic produces a new challenging issue that how effectively shares containers among tasks to avoid redundant container image download processes and redundant task allocations. One of the fundamental problems is that no policy has been established for simultaneously satisfying effective container sharing, maintaining the degree of task parallelism, and effective computational resource utilization. In this paper, we propose a clustering-based containerized task scheduling algorithm for clouds, namely, shareable functional task clustering for utilizing virtualized resources (SF-CUV). The objective of SF-CUV is to minimize the makespan with less computational resources and containers than other algorithms by clustering tasks and sharing each container among tasks. SF-CUV consists of two phases: (i)task clustering and pre-virtual CPU (vCPU) allocation phase to derive an accurate scheduling priority, and (ii)task ordering and actual task reallocation phase. Experimental results obtained via simulation and in a real environment show that SF-CUV can utilize both vCPUs and containers with a shorter makespan compared with other approaches. [ABSTRACT FROM AUTHOR]

Multiple tasks arrive in the distributed systems that can be executed in either parallel or sequential manner. Before the execution, tasks are scheduled prioritywise with the help of scheduling algorithms to their respective processors. For task assignment, every scheduling algorithm follows different protocols like upper bound of CPU utilization, assigning priorities, etc. In this paper, author has worked on such scheduling algorithms. Previously, the author evaluated the performance of algorithms on the basis of transactions (group of tasks). In this paper, the author re-evaluates joint EDF–RM scheduling algorithm, where its performance is calculated on the execution of individual task basis. For comparative analysis, similar algorithms are considered, i.e., joint EDF–RMS, earliest deadline first (EDF) and rate monotonic scheduling (RMS). These mentioned algorithms are simulated and analyzed with the help of statistical analysis, and turnaround time of periodic tasks is evaluated. Additionally, migration distribution and CPU utilization on the basis of scheduling algorithms' upper bounds are also calculated. [ABSTRACT FROM AUTHOR]

SRAM-based FPGAs feature high performance and flexibility. Thus, they have found many applications in modern high-performance computing (HPC) systems. These systems suffer from the limitation of the computing resources problem for running HPC applications. Therefore, multi-FPGA systems have been emerged to alleviate such resource limitations. In this regard, efficient scheduling strategies are required to dynamically steer the execution of applications—represented as task graphs—on a set of connected FPGAs. In this paper, a heuristic-based dynamic critical path-aware scheduling technique named CPA is presented to schedule task graphs on multi-FPGA systems. The proposed technique, by considering the computation and communication capabilities of FPGAs, dynamically assigns priority to tasks in different steps in order to achieve better makespans. The proposed technique has been evaluated by conducting several experiments on real-world and three different shapes of random task graphs with different number of tasks, and its efficiency has been compared with that of three task graph scheduling approaches. The obtained results demonstrate that the proposed CPA technique outperforms well-known heuristic scheduling strategies and improves their makespan by 13.47% on average. In addition, the experiments show that the proposed technique generates the schedules in the order of milliseconds and the average of its yielded makespans is 12.05% longer than that of an optimum schedule. [ABSTRACT FROM AUTHOR]

In partially run-time reconfigurable (PRR) FPGAs, hardware tasks should be configured before their execution. The configuration delay imposed by the reconfiguration process increases the total execution time of the hardware tasks and task graphs. In this paper, a new technique named forefront-fetch is presented to improve the makespan of hardware task graphs running on PRR FPGAs via alleviating the adverse effects of the configuration delays. In this technique, which is applied to a sequence of task graphs, the configuration of some tasks is carried out within the execution phase of the previous task graph. This strategy leads to hide the configuration delay of the forefront-fetched tasks that as a result improves the execution time. The proposed solution modifies the schedules of the task graphs at design time to obtain a set of schedule pairs for the run-time environment. Experiments on actual and synthesized task graphs demonstrate the ability of the proposed technique in improving the makespan of hardware task graphs. The obtained results show that for a set of task graphs running on Xilinx™ Virtex-5 XUPV5LX110T FPGA, makespan is improved by 37.81% on average. Moreover, the proposed solution outperforms state-of-the-art prefetch-aware scheduling strategies by 14.78% makespan improvement. [ABSTRACT FROM AUTHOR]

Reconfigurable devices (RDs) are extremely advantageous when employed in real-time embedded systems. Nonetheless, they are susceptible to soft errors. In a broad sense, the present research addresses the challenge of improving the reliability of independent periodic real-time hardware tasks in RDs by utilizing hybrid fault-tolerant scheduling. The current paper combines static and dynamic real-time scheduling techniques to improve the reliability of the system. First, the proposed algorithm statically schedules primary tasks and preserves area and time for possible backup tasks on the RD. The overlapping of passive backup tasks is possible. Next, at the run time, event-triggered dispatcher dynamically determines which candidate backup copy should be selected for configuration on the overloaded preserved areas. Reliability, task deadline, and RD area limitations are the determining factors of backup overloading in the static phase. On the other hand, in the dynamic phase, the execution result of the primary tasks—in this case, success or failure—is the deciding factor based on which the dispatcher configures the true backup task on the preserved area. Experimental results show that the hybrid scheduling technique enhances the mean-time-to-failure of the system by an average factor of 1.22 in comparison with a similar state-of-the-art study. [ABSTRACT FROM AUTHOR]

Efficient task and workflow scheduling are very crucial for increasing performance, resource utilization, customer satisfaction, and return of investment for cloud service providers. Based on the number of clouds that the scheduling schemes can support, they can be classified as single-cloud and inter-cloud scheduling schemes. This paper presents a comprehensive survey and an overview of the inter-cloud scheduling schemes aimed to allocate user-submitted tasks and workflows to the appropriate virtual machines on multiple clouds regarding various objectives and factors. It classifies the scheduling schemes designed for a variety of inter-cloud environments and describes their architecture, key features, and advantages. Also, the inter-cloud scheduling approaches are compared and their various features are highlighted. Finally, the concluding remarks and open research challenges in the multi-cloud scheduling context are illuminated. [ABSTRACT FROM AUTHOR]

Efficient scheduling of tasks in heterogeneous computing systems is of primary importance for high-performance execution of programs. The programs are to be considered as multiple sequences of tasks that are presented as directed acyclic graphs (DAG). Each task has its own execution timeline that incorporates into multiple processors. Moreover, each edge on the graph represents constraints between the sequenced tasks. In this paper, we propose a new list-scheduling algorithm that schedules the tasks represented in the DAG to the processor that best minimizes the total execution time by taking into consideration the restriction of crossover between processors. This objective will be achieved in two major phases: (a) computing priorities of each task that will be executed, and (b) selecting the processor that will handle each task. The first phase, priorities computation, focuses on finding the best execution sequence that minimizes the makespan of the overall execution. In list-scheduling algorithm, the quality of the solution is very sensitive to the priority assigned to the tasks. Therefore, in this paper, we include an enhanced calculation of weight that is used in the ranking equation for determining the priority of tasks. The second phase, processor selection, primarily focuses on allocating a processor that is a best fit for each task to be executed. In this paper, we enhance the processor selection by introducing a randomized decision mechanism based on a threshold which decides whether the task be assigned to the processor with the lowest execution time or to the processor that produces the lowest finish time. This mechanism considers a balanced combination of the local and global optimal results to explore the search space efficiently to optimize the overall makespan. The proposed algorithm is evaluated on different randomly generated DAGs, and results are compared with the well-known existing approaches to show the effectiveness of the proposed algorithm in reducing the makespan of execution. The experiment's results show improvement in the makespan that reaches up to 6-7%. [ABSTRACT FROM AUTHOR]

Service-level agreement (SLA) is a major issue in cloud computing because it defines important parameters such as quality of service, uptime, downtime, period of service, pricing, and security. However, the service may vary from one cloud service provider (CSP) to another. The collaboration of the CSPs in the heterogeneous multi-cloud environment is very challenging, and it is not well covered in the recent literatures. In this paper, we present two SLA-based task scheduling algorithms, namely SLA-MCT and SLA-Min-Min for heterogeneous multi-cloud environment. The former algorithm is a single-phase scheduling, whereas the latter one is a two-phase scheduling. The proposed algorithms support three levels of SLA determined by the customers. Furthermore, the algorithms incorporate the SLA gain cost for the successful completion of the service and SLA violation cost for the unsuccessful end of the service. We simulate the proposed algorithms using benchmark and synthetic datasets. The experimental results of the proposed SLA-MCT are compared with three single-phase task scheduling algorithms, namely CLS, Execution-MCT, and Profit-MCT, and the results of the proposed SLA-Min-Min are compared with two-phase scheduling algorithms, namely Execution-Min-Min and Profit-Min-Min in terms of four performance metrics, namely makespan, average cloud utilization, gain, and penalty cost of the services. The results clearly show that the proposed algorithms properly balance between makespan and gain cost of the services in comparison with other algorithms. [ABSTRACT FROM AUTHOR]