Resource Scheduling Algorithm for Heterogeneous High-Performance Clusters

The planning and deployment of mega constellations has posed increasing pressure on the existing telemetry, track and command (TTC) resource scheduling strategy. The current plan-based scheduling strategy distributes resources according to needs, however, it cannot address sudden resource requests with prompt responses. By analyzing the status quo of resource scheduling systems, we have proposed a distributed collaborative scheduling technology for random access in real-time, based on the theory of low costs and rapid transfers of scheduling task resources. First, we use knowledge graph technology to map the link between scheduling tasks and idle resources and develop a graph network. Furthermore, by analyzing and evaluating the resource transfer capability of a multi-target TTC system in hemispherical coverage (MT-TTC), we propose the concept of a virtual resource pool and develop a two-layer scheduling framework in which the central scheduling system coordinates with multiple regional systems. This scheduling strategy works well when a sudden TTC request occurs. The central system manages the satellite requests and make plans, while regional systems are responsible for handling the sudden requests and report to the central system on its resource change. With regional systems sharing some degree of autonomy, the central system making a final decision and systems across the framework coordinating in real-time, we design a model on resource transfers for MT-TTC equipment and a collaboration mechanism that coordinates the resource transfers across MT-TTC equipment and other single-beam equipment within and between regions. We do experiments in simulation scheduling environments. The results show that the regional scheduling system based on MT-TTC equipment is able to manage continuous random-access requests in real-time, and avoid resource conflicts across regions by collaborating with the central scheduling system. It is of value to be applied in real scenarios.

Inverse synthetic-aperture radar (ISAR) can achieve precise imaging of targets, which enables precise perception of battlefield information, and it has become one of the most important tasks for radar systems. In multi-target scenarios, a resource scheduling method is required to improve the sensing ability and the overall efficiency of a radar system due to the limited resources. Considering the motion state of the target will change as the observation distance increases and image defocusing can occur due to the prolonged coherence accumulation time and significant changes in the target’s motion state, the optimal observation period should be an important consideration factor in the resource scheduling method to further improve the imaging efficiency of radar system, which has not yet been involved in existing research. In this paper, we first derive the expressions of the target’s effective rotation angle and the equivalent rotation angular velocity and then define the target’s optimal observation period. Then, for multi-target imaging scenarios, we allocate pulse resources within a given time period based on sparse-aperture ISAR imaging technology. An adaptive radar resource scheduling algorithm for multi-target ISAR imaging is proposed, which prioritizes allocating resources based on the optimal observation periods for the targets. In the algorithm, a radar resource scheduling model for multi-target ISAR imaging is established, and a feedback-based closed-loop search optimization method is proposed to solve the model. Finally, the best scheduling strategy can be obtained, which includes imaging task duration and the pulse allocation sequence for each target. Simulation results validate the effectiveness of the algorithm.

Deep learning (DL) is gaining popularity in various fields, such as computer vision, speech identification, and self-driving vehicles. Blackwell graphical processing units (GPUs) and Grace central processing units (CPUs) are widely utilized in computing platforms, including portable gadgets and supercomputers. Optimizing performance for multiple-programming workloads on such devices is crucial, as a single application cannot typically utilize all available resources. Developing effective GPU resource administration and scheduling systems is crucial for effective distribution among several clients. Nevertheless, current versions have several flaws. Clients often provide wrong task resource requirements, resulting in cluster resource low utilization. DL-based resource scheduling without considering cluster network features leads to poor operational efficiency. To address these difficulties, this study develops a distributed data center (DDC) framework using Blackwell GPUs and Grace CPUs for resource scheduling and accelerated computing with DL. This study offers a regression model-based technique for estimating task resource requirements, preventing clients from overallocating computer resources. This paper demonstrates smart cluster network-effective scheduling approaches for both prompt and batch forms based on the resource need assessment methodologies described. Study findings demonstrate that the suggested system can improve task computation by 19% overall and reduce mean job completion duration (JCD) by 22% when evaluated against modern approaches. Furthermore, the presented optimization approaches are useful in a variety of contexts.

Besides avoiding inter-symbol interference and leading to high capacity, wireless orthogonal frequency division multiplexing (OFDM) provide fine granularity for resource allocation since they are capable of dynamically assigning sub-carriers to multiple users and adaptively allocating transmit power. The current dominate layered networking architecture, in which each layer is designed and operated independently, results in inefficient and inflexible resource use in wireless networks due to the nature of the wireless medium, such as time-varying channel fading. Thus, we need an integrated adaptive design across different layers. In this paper, we focus on resource allocation and scheduling in wireless multiuser OFDM networks based on joint physical and medium access control (MAC) layer optimization. An adaptive cross-layer design for the downlink multiuser OFDM systems, to maximize the weighted sum capacity of all users, where each user has multiple heterogeneous traffic queues simultaneously is proposed. A packet dependent (PD) scheduling scheme is employed at the MAC layer, which determines the packet transmission order by assigning different weights to different packets, and is shown by simulations more efficient than the previous methods where all packets in a queue have the same weight. The weight design in PD scheduling considers the delay, size and quality of service (QoS) priority level of packets. Each user weight employed in resource allocation at the physical (PHY) layer is obtained by summing up the weights of selected packets for the user. We also deeply investigate the various resource scheduling schemes for comparisons.

Resource scheduling and performance analysis of hybrid renewable energy systems with carbon neutrality consideration: A scenario-based multi-agent approach

The purpose of resource scheduling is to deal with all kinds of unexpected events that may occur in life, such as fire, traffic jam, earthquake and other emergencies, and the scheduling algorithm is one of the key factors affecting the intelligent scheduling system. In the traditional resource scheduling system, because of the slow decision-making, it is difficult to meet the needs of the actual situation, especially in the face of emergencies, the traditional resource scheduling methods have great disadvantages. In order to solve the above problems, this paper takes emergency resource scheduling, a prominent scheduling problem, as an example. Based on Vague set theory and adaptive grid particle swarm optimization algorithm, a multi-objective emergency resource scheduling model is constructed under different conditions. This model can not only integrate the advantages of Vague set theory in dealing with uncertain problems, but also retain the advantages of adaptive grid particle swarm optimization that can solve multi-objective optimization problems and can quickly converge. The research results show that compared with the traditional resource scheduling optimization algorithm, the emergency resource scheduling model has higher resolution accuracy, more reasonable resource allocation, higher efficiency and faster speed in dealing with emergency events than the traditional resource scheduling model. Compared with the conventional fuzzy theory emergency resource scheduling model, its handling speed has increased by more than 3.82 times.

Visible light communication (VLC) has drawn much attention in the field of high-rate indoor wireless communication. While most existing works focused on point-to-point VLC technologies, few studies have concerned multiuser VLC, where multiple optical access points (APs) transmit data to multiple user receivers. In such scenarios, inter-user interference constitutes the major factor limiting the system performance. Therefore, a proper scheduling scheme has to be proposed to coordinate the interference and optimize the whole system performance. In this work, we aim to maximize the sum rate of the system while taking into account user fairness by appropriately assigning LED lamps to multiple users. The formulated scheduling problem turns out to be a maximum weighted independent set problem. We then propose a novel and efficient resource allocation method based on graph theory to achieve high sum rates. Moreover, we also introduce proportional fairness into our scheduling scheme to ensure the user fairness. Our proposed scheduling scheme can, with low complexity, achieve more multiplexing gains, higher sum rate, and better fairness than the existing works.

Installing multiple antennas at user equipments (UEs) may pose great challenges. In virtual multiple-input multiple-output (MIMO) concept, multiple single-antenna UEs can be grouped to transmit signals at the same time. Their signals can be separated at the base station (BS) by the same methods already used in conventional point-to-point MIMO scenarios. An important problem in virtual MIMO application is which users can be scheduled simultaneously to achieve a good balance between various system targets, e.g. aggregate throughput and user fairness. In this paper, a novel scheduling scheme is proposed which combines the advantages of proportional fair and maximum rate rules, based on successive interference cancellation (SIC) MIMO receiver architecture. Simulation results show that the proposed scheme can achieve a good tradeoff between system throughput and user fairness. Compared with other scheduling schemes already proposed in virtual MIMO context, e.g., orthogonality pairing scheduling (OPS), the proposed scheme can be justified by its low computational complexity and flexibility.

With the rapid development of the mobile Internet, smart devices are playing an increasingly important role in people's lives. Edge computing is a key technology for solving resource scheduling problems and is of great significance for meeting high-performance user requirements. This paper deeply studies the resource scheduling strategy for Internet of Things edge computing based on artificial potential field algorithm. It analyzes the architecture of edge computing, resource scheduling methods, research problems, and technical methods, and discusses its application scenarios and challenges. Through the introduction of artificial potential field algorithm, the edge computing resource scheduling is optimized to improve system performance and user experience.

In a 5G ultra-dense network, dynamic network topology and traffic patterns lead to excessive system overhead and complex radio resource conflicts. The cloud radio access network and the fog computing have the advantages of high computation capabilities and low transmission delays. Therefore, by taking full advantage of these two characteristics, this study proposes a novel radio resource coordination and scheduling scheme in an ultra-dense cloud-based small cell network. Interference among small cells (or remote radio heads) can be avoided by implementing centralized cooperative processing in the base band unit pool in advance. Resource sharing in coordination and transfer depend on fog computing to relieve the overloaded cloud processing platform and reduce transmission delays, thereby maximizing resource utilization and minimizing system overhead when the network topology and number of users change dynamically. The simulation shows that the proposed scheme can increase the system throughput by 20% compared with the clustering-based algorithm; it can also increase system throughput by 33% compared with the graph coloring algorithm, decrease the signaling overhead by about 50%, and improve network’s quality of service.

Clusters of computers have emerged as mainstream parallel and distributed platforms for high‐performance, high‐throughput and high‐availability computing. To enable effective resource management on clusters, numerous cluster management systems and schedulers have been designed. However, their focus has essentially been on maximizing CPU performance, but not on improving the value of utility delivered to the user and quality of services. This paper presents a new computational economy driven scheduling system called Libra, which has been designed to support allocation of resources based on the users' quality of service requirements. It is intended to work as an add‐on to the existing queuing and resource management system. The first version has been implemented as a plugin scheduler to the Portable Batch System. The scheduler offers market‐based economy driven service for managing batch jobs on clusters by scheduling CPU time according to user‐perceived value (utility), determined by their budget and deadline rather than system performance considerations. The Libra scheduler has been simulated using the GridSim toolkit to carry out a detailed performance analysis. Results show that the deadline and budget based proportional resource allocation strategy improves the utility of the system and user satisfaction as compared with system‐centric scheduling strategies. Copyright © 2004 John Wiley & Sons, Ltd.

: The Range Commanders Council (RCC), specifically the Data Reduction and Computing Group (DR&CG), recognized the need for an integrated Range Resource Management and Scheduling capability that would facilitate support of all phases of test or training events. Such a tool would span the initial planning phases, the execution and evaluation of an exercise, and finally, the accounting and billing process. Currently, each range not only performs its functions differently, but the tools, systems, and steps followed are also unique to each facility. This study was initiated with the goal of providing the ranges with a document describing the areas to consider when designing and implementing such an integrated tool. A three-step approach was used in the conduct of this study First, documentation was solicited from the different RCC members who represent the DoD ranges. Information was gathered that focused on the functionality and capability of current range resource management and scheduling systems. Interviews with subject matter experts complemented the study when actual written material was not available. Second, the information from each individual range was analyzed, organized, and then compared with the other ranges. Actual requirements were identified from systems documentation or from interviews and generalized in order to present a broader definition of required capabilities. These requirements include the processes, types of data, design documentation, system architecture, and user interfaces in current use. Finally, a recommended approach for use in developing a resource management and scheduling tool was defined that takes into consideration future requirements as well as current systems development and design practices.

Edge computing is a domain that processes extensive data by integrating data processing and storage at one device. The vast improvement in resource deployment for edge environments is the primary reason for the difficulties in resource scheduling and security management. In this paper, a Novel Edge Computing Resource Scheduling (NECRS) system, such as the Lightweight Security-Aware Resource Scheduler (LSARS), is presented. The proposed LSARS logically distributes computing resources and assesses the security risk, based on the trust measure, data sensitivity, and node vulnerability. The proposed NECRS utilizes a hybrid decision-making approach that integrates a heuristic-based scheduling technique with lightweight anomaly detection to provide low latency and resilient task execution times. Extensive experimental results demonstrate that LSARS outperforms conventional scheduling schemes in terms of resource utilization, scheduling efficiency, and attack mitigation. Finally, the proposed approach improved the security and responsiveness of edge-enabled services in real-world deployments.

In order to resolve resource management and scheduling problem in Manufacturing Grid (MG) – an application of Grid technology, we develop a resource management and scheduling system with the interaction of Manufacturing Grid Information Service (MGIS) and the Manufacturing Grid Resource Scheduler (MGRS). The former, MGIS, provides fundamental mechanisms for remote resource encapsulating, registration, and monitoring, and the latter, MGRS, performs scheduling roles as Global Process Planning (GPP) analyzing, resource discovery, resource selection, and resource mapping.

Currently, the scheduling mode used by most TT&C resource scheduling system was based on the mission requirements offered by users. After the scheduling plan is formulated, the resource redundancy has been basically determined, without the advance planning of the resources. This article analyzed the data of the scheduling system, by dividing the requirements of resources into two types of “period” and “peak”. Through analyzing the periodic use of resource historical data, the periodic resource demand will be predicted and guided resources for time-dividing and domain-planning. By analyzing the allocation data of dynamic resource scheduling, the peak resource will be reserved to enable and recycle dynamically in actual. Finally the strategy was verified by simulation, which provided a different approach to improve the utilization rate of the entire network resources.KeywordsData miningPeriodic resourcePeak resource