Showing papers on "Task (computing) published in 2022"

Integrated Scheduling of Real-Time and Interactive Tasks for Configurable Industrial Systems

Abstract: With the recent advances in Internet of Things and cyber-physical systems technologies, smart industrial systems support configurable processes consisting of human interactions as well as hard real-time functions. This implies that irregularly arriving interactive tasks and traditional hard real-time tasks coexist. As the characteristics of the tasks are heterogeneous, it is not an easy matter to schedule them all at once. To cope with this situation, this article presents a new task scheduling policy that uses the notion of “virtual real-time task” and two-phase scheduling. As hard real-time tasks must keep their deadlines, we perform offline scheduling based on genetic algorithms beforehand. This determines the processor's voltage level and memory location of each task and also reserves the virtual real-time tasks for interactive tasks. When interactive tasks arrive during the execution, online scheduling is performed on the time slot of the virtual real-time tasks. As interactive workloads evolve over time, we monitor them and periodically update the offline scheduling. Experimental results show that the proposed policy reduces the energy consumption by 66.8% on average without deadline misses and also supports the waiting time of less than 3 (s) for interactive tasks.

Design and Performance Characterization of RADICAL-Pilot on Leadership-Class Platforms

Abstract: Many extreme scale scientific applications have workloads comprised of a large number of individual high-performance tasks. The Pilot abstraction decouples workload specification, resource management, and task execution via job placeholders and late-binding. As such, suitable implementations of the Pilot abstraction can support the collective execution of large number of tasks on supercomputers. We introduce RADICAL-Pilot (RP) as a portable, modular and extensible pilot-enabled runtime system. We describe RP's design, architecture and implementation. We characterize its performance and show its ability to scalably execute workloads comprised of tens of thousands heterogeneous tasks on DOE and NSF leadership-class HPC platforms. Specifically, we investigate RP's weak/strong scaling with CPU/GPU, single/multi core, (non)MPI tasks and Python functions when using most of ORNL Summit and TACC Frontera. RADICAL-Pilot can be used stand-alone, as well as the runtime for third-party workflow systems.

Efficient, Dynamic Multi-Task Execution on FPGA-Based Computing Systems

Abstract: With growing Field Programmable Gate Array (FPGA) device sizes and their integration in environments enabling sharing of computing resources such as cloud and edge computing, there is a requirement to share the FPGA area between multiple tasks. The resource sharing typically involves partitioning the FPGA space into fix-sized slots. This results in suboptimal resource utilisation and relatively poor performance, particularly as the number of tasks increase. Using OpenCL’s exploration capabilities, we employ clever clustering and custom, task-specific partitioning and mapping to create a novel, area sharing methodology where task resource requirements are more effectively managed. Using models with varying resource/throughput profiles, we select the most appropriate distribution based on the runtime, workload needs to enhance temporal compute density. The approach is enabled in the system stack by a corresponding task-based virtualisation model. Using 11 high performance tasks from graph analysis, linear algebra and media streaming, we demonstrate an average $2.8\times$ 2 . 8 × higher system throughput at $2.3\times$ 2 . 3 × better energy efficiency over existing approaches.

SLA-Based Group Tasks Max-Min (GTMax-Min) Algorithm for Task Scheduling in Multi-Cloud Environments

Abstract: Cloud is a distributed heterogeneous computing paradigm that facilitates on-demand delivery of IT heterogeneous resources to the customer based on their needs over the Internet with a pay-as-per service they use. Service level agreement (SLA) specifies the customer’s expected service levels through cloud service provider (CSP) and the remedies or penalties if any of the CSP does not meet agreed-on service levels. Before providing the requested services to the customer, CSP and customer negotiate and sign on an SLA. CSP earns money for the service provided to the customer on satisfying the agreed-on service levels. Otherwise, CSP pays the penalty cost to the customer for the violation of SLA. Task scheduling minimizes task execution time and maximizes resource usage rate. Scheduling objective tends to improve quality of service (QoS) parameters like resource usage, with a minimum execution time and cost (without violating SLA). The proposed algorithm SLA-GTMax-Min schedules the tasks efficiently to the heterogeneous multi-cloud environment satisfying SLA and balances makespan, gain, and penalty/violation cost. Proposed SLA-GTMax-Min represents three levels of SLA provided with three types of services expected by the customers. The services are namely tasks minimum execution time, tasks minimum gain cost, and tasks both minimum execution time and gain cost in percentage, respectively. Makespan is termed as tasks minimum execution time. Gain cost represents minimum execution cost for completing tasks execution. The proposed algorithm SLA-GTMax-Min incorporates the SLA gain cost for providing service successfully and SLA violation cost for providing service unsuccessfully. Performance analysis of algorithm SLA-GTMax-Min and existing algorithm is measured based on the benchmark dataset values. The experimental results of SLA-GTMax-Min algorithm and the existing scheduling algorithms, namely, SLA-MCT, Execution-MCT, Profit-MCT, SLA-Min-Min, Execution-Min-Min, and Profit-Min-Min, are compared by evaluation metrics. Evaluation measure considered for evaluating the performance of the proposed SLA-GTMax-Min algorithm are makespan, cloud utilization ratio, gain cost is the cost earned by the CSP for successful completion of the tasks, and penalty cost the CSP pays to the customer for violation of SLA. The experimental results illustrate clearly algorithm SLA-GTMax-Min performs a better balance among makespan, gain cost, and penalty cost than existing algorithms.

Comparative Study and Detection of COVID-19 and Related Viral Pneumonia Using Fine-Tuned Deep Transfer Learning

Abstract: Coronavirus (or COVID-19), which came into existence in 2019, is a viral pandemic that causes illness and death in the lives of human. Relentless research efforts have been on to improve key performance indicators for detection, isolation and early treatment. The aim of this study is to conduct a comparative study on the detection of COVID-19 and develop a Deep Transfer Learning Convolutional Neural Network (DTL-CNN) Model to classify chest X-ray images in a binary classification task (as either COVID-19 or Normal classes) and a three-class classification scenario (as either COVID-19, Viral-Pneumonia or Normal categories). Dataset was collected from Kaggle website containing a total of 600 images, out of which 375 were selected for model training, validation and testing (125 COVID-19, 125 Viral Pneumonia and 125 Normal). In order to ensure that the model generalizes well, data augmentation was performed by setting the random image rotation to 15 degrees clockwise. Two experiments were performed where a fine-tuned VGG-16 CNN and a fine-tuned VGG-19 CNN with Deep Transfer Learning (DTL) were implemented in Jupyter Notebook using Python programming language. The system was trained with sample datasets for the model to detect coronavirus in chest X-ray images. The fine-tuned VGG-16 and VGG-19 DTL models were trained for 40 epochs with batch size of 10, using Adam optimizer for weight updates and categorical cross entropy loss function. A learning rate of 1e−2 was used in fine-tuned VGG-16 while 1e−1 was used in fine-tuned VGG-19, and was evaluated on the 25% of the X-ray images. It was discovered that the validation and training losses were significantly high in the earlier epochs and then noticeably decreases as the training occurs in more subsequent epochs. Result showed that the fine-tuned VGG-16 and VGG-19 models, in this work, produced a classification accuracy of 99.00% for binary classes, and 97.33% and 89.33% for multi-class cases respectively. Hence, it was discovered that the VGG-16 based DTL model classified COVID-19 better than the VGG-19 based DTL model. Using the best performing fine-tuned VGG-16 DTL model, tests were carried out on 75 unlabeled images that did not participate in the model training and validation processes. The proposed models, in this work, provided accurate diagnostics for binary classification (COVID-19 and Normal) and multi-class classification (COVID-19, Viral Pneumonia and Normal), as it outperformed other existing models in the literature in terms of accuracy.

Harnessing the Potential of Function-Reuse in Multimedia Cloud Systems

Abstract: Cloud-based computing systems can get oversubscribed due to the budget constraints of their users or limitations in certain resource types. The oversubscription can, in turn, degrade the users perceived Quality of Service (QoS). The approach we investigate to mitigate both the oversubscription and the incurred cost is based on smart reusing of the computation needed to process the service requests (i.e., tasks). We propose a reusing paradigm for the tasks that are waiting for execution. This paradigm can be particularly impactful in serverless platforms where multiple users can request similar services simultaneously. Our motivation is a multimedia streaming engine that processes the media segments in an on-demand manner. We propose a mechanism to identify various types of “mergeable” tasks and aggregate them to improve the QoS and mitigate the incurred cost. We develop novel approaches to determine when and how to perform task aggregation such that the QoS of other tasks is not affected. Evaluation results show that the proposed mechanism can improve the QoS by significantly reducing the percentage of tasks missing their deadlines and reduce the overall time (and subsequently the incurred cost) of utilizing cloud services by more than 9 percent.

Improvement of Task Offloading for Latency Sensitive Tasks in Fog Environment

Abstract: Fog computing is gaining rapid acceptance as a distributed computing paradigm that brings cloud-like services near the end devices. It enhances the computation capabilities of mobile nodes and IoT (Internet of Things) devices by providing compute and storage capabilities similar to the cloud but at a lower latency and using lesser bandwidth. Additional advantages of fog computing include its support for node mobility, context awareness, reliability and scalability. Due to its multiple benefits, fog computing is used for offloading tasks from applications executing on end devices. This allows faster execution of applications using the capabilities of fog nodes. However, the task offloading problem in the fog environment is challenging due to the dynamic nature of fog environment and multiple QoS (Quality of Service) parameters dependent on the application being executed. Therefore, the chapter proposes a QoS-aware task offloading strategy using a novel nature-inspired optimization algorithm, known as the Smart Flower Optimization Algorithm (SFOA). The proposed strategy takes into account the QoS parameters such as the task deadlines and budget constraints in selection of appropriate fog nodes where computation tasks can be offloaded. The proposed strategy has been simulated and the results have verified the efficacy of the strategy.

Real-Time Systems

Abstract: We come across several times the term “real time” tagged before any other noun or verb, like real-time data, real-time monitoring, real-time governance, and so on. Let us understand what real time signifies. After completing this chapter, one will be able to un-tag the term “real time” from many such usages. In this chapter, we will understand the characteristics of a system that qualifies it, to be called as real-time system. Then, we will classify the RT systems based on their traits. We will study the reference model by which we can analyze the system and focus on important aspects of them. We will study scheduling mechanisms through supporting algorithms to reach real-time constraints. Section 6.2 classifies real-time systems (RTS) to periodic, mostly periodic, aperiodic, and predictable and unpredictable (spurious) systems. Section 6.4 deals with models to execute such periodic tasks. Section 6.6 classifies scheduling algorithms. Section 6.7 deals with clock-driven scheduling. Section 6.8 deals with scheduling priority-driven periodic tasks. Section 6.9 deals with scheduling tasks with dynamic priority like Earliest Deadline First (EDF) and Least Slack Time First (LST). Section 6.10 deals with scheduling sporadic tasks. Section 6.11 deals with accessing resources by multiple tasks, handling the contention for resources and how to handle cases of priority inversion. To summarize, aperiodic jobs are soft and can be accommodated by stealing slack times and idle slots. Tasks can be prioritized based on their rate. RMA is a popular protocol. Priorities of jobs can be assigned using early deadlines and also the least slack time. EDF algorithms are most popular. Sporadic jobs are unpredictable with varied properties. Given a context, a sporadic job can be accepted if it is schedulable. Sporadic jobs have to be handled in a separate queue. The above algorithms assume no contention of resources. Resource contention modifies the execution times based on the availability of resources and the critical section of the resources in each job. The most serious problem is priority inversion, which has to be taken care with multiple algorithms like priority inheritance. This chapter becomes the input to the next chapter where we study the architecture of real-time executives, their standardization, and their features.