Highly reliable matching method based on multi-dimensional resource measurement and rescheduling in computing power network

doi:10.11772/j.issn.1001-9081.2024111653

Journal of Computer Applications ›› 2025, Vol. 45 ›› Issue (11): 3632-3641.DOI: 10.11772/j.issn.1001-9081.2024111653

• Advanced computing • Previous Articles

Highly reliable matching method based on multi-dimensional resource measurement and rescheduling in computing power network

Lin WEI¹, Jinyang LI¹, Yajie WANG², Mengyang HE¹()

^1.School Cyber Science and Engineering，Zhengzhou University，Zhengzhou Henan 450002，China
^2.National Computer Network Emergency Response Technical Team/Coordination Center of China，Beijing 100029，China

Received:2024-11-27 Revised:2025-06-11 Accepted:2025-06-12 Online:2025-06-23 Published:2025-11-10
Contact: Mengyang HE
About author:WEI Lin， born in 1968， M. S.， associate professor. Her research interests include network and distributed computing， data science and intelligent computing， information security.
LI Jinyang， born in 1999， M. S. candidate. His research interests include computing power network， edge computing.
WANG Yajie， born in 1992， M. S.， engineer. His research interests include information security， artificial intelligence.
Supported by:
Science and Technology Research Project of Henan Province(232102210154);Provincial Science and Technology Program of Henan Province(YYJC022022001);Major Science and Technology Special Project of Henan Province(241110210200)

算力网络中基于多维资源度量和重调度的高可靠匹配方法

卫琳¹, 李金阳¹, 王亚杰², 和孟佯¹()

^1.郑州大学网络空间安全学院，郑州 450002
^2.国家计算机网络应急技术处理协调中心，北京 100029

通讯作者: 和孟佯
作者简介:卫琳（1968—），女，河南郑州人，副教授，硕士，主要研究方向：网络与分布式计算、数据科学与智能计算、信息安全
李金阳（1999—），男，河南周口人，硕士研究生，主要研究方向：算力网络、边缘计算
王亚杰（1992—），男，北京人，工程师，硕士，主要研究方向：信息安全、人工智能
基金资助:
河南省科技攻关项目(232102210154);河南省省级科技计划项目(YYJC022022001);河南省重大科技专项(241110210200)

Abstract

Abstract:

Computing Power Network （CPN） is a new network system that solves the contradiction between computing power supply and demand， network transmission problems， and the issue of universal access to computing resources. According to the supply capacity of computing power resource providers and the dynamic resource requirements of application demanders， the computing， storage， network and other multi-dimensional resources of the underlying computing power infrastructure in the region are integrated to provide users with personalized computing power resource services and realize efficient management and on-demand allocation of computing power resources. To enhance the utilization and reliability of CPN resource matching and scheduling， a highly reliable matching method was proposed， namely Resource Measurement and Rescheduling Matching Method （RMRMM）. To achieve high-utilization resource scheduling， RMRMM designed a resource measurement matching scheme based on entropy weighted Technique for Order Preference by Similarity to Ideal Solution （entropy weighted TOPSIS） method and Deep Reinforcement Learning （DRL）， comprehensively measured the Structural Feature Value （SFV）， computing power， storage capacity， and network communication capacity of the node， and narrowed the resource matching range to improve the matching accuracy and resource utilization. Additionally， RMRMM considered the failure of nodes due to attacks， and designed a rescheduling module based on the Adaptive Large Neighborhood Search （ALNS） algorithm. When matches failed， nodes and tasks were rescheduled to improve the acceptance rate of tasks and enhance the overall reliability. Simulation experimental results on OMNet++ platform demonstrate that average BandWidth （BW） utilization， average Random Access Memory （RAM） utilization， average STORAGE utilization， and task request reception rate of RMRMM reach 69.7%， 66.4%， 68.5%， and 75.5%， respectively. Both resource utilization and request reception rate of RMRMM outperform other matching strategies， improving the efficiency and reliability of RMRMM.

Key words: Computing Power Network (CPN), resource measurement, rescheduling, high reliability, Deep Reinforcement Learning (DRL)

摘要：

算力网络（CPN）是一种解决算力供需矛盾、网络传输问题以及算力资源普惠问题的新型网络体系，根据算力资源提供方的供给能力和应用需求方的动态资源需求，对区域内算力基础设施底层的计算、存储、网络等多维资源进行整合，为用户提供个性化的算力资源服务，实现算力资源的高效管理和按需分配。为了提高CPN资源匹配调度的利用率和可靠性，提出一种基于多维资源度量和重调度的高可靠匹配方法（RMRMM）。为了实现高利用率的资源调度，RMRMM设计了基于熵权优劣解距离法（entropy weighted TOPSIS）和深度强化学习（DRL）的资源度量匹配方案，对节点的结构特征值（SFV）、计算能力、存储能力、网络通信能力进行综合度量，缩小资源匹配范围以提高匹配的精准性和资源的利用率；同时RMRMM考虑节点遭受攻击失效的情况，设计基于自适应大邻域搜索（ALNS）算法的重调度模块，在匹配结果失效时进行节点与任务的重新调度，提高任务的接收率以增强整体的可靠性。在OMNet++平台上的仿真实验结果表明，RMRMM的平均带宽（BW）利用率、平均主存（RAM）利用率、平均存储（STORAGE）利用率和任务请求接收率最高达到69.7%、66.4%、68.5%、75.5%，资源利用率和任务请求接收率均优于其他匹配策略，说明RMRMM更加高效可靠。

关键词: 算力网络, 资源度量, 重调度, 高可靠, 深度强化学习

CLC Number:

TP393.07

Lin WEI, Jinyang LI, Yajie WANG, Mengyang HE. Highly reliable matching method based on multi-dimensional resource measurement and rescheduling in computing power network[J]. Journal of Computer Applications, 2025, 45(11): 3632-3641.

卫琳, 李金阳, 王亚杰, 和孟佯. 算力网络中基于多维资源度量和重调度的高可靠匹配方法[J]. 《计算机应用》唯一官方网站, 2025, 45(11): 3632-3641.

Figures/Tables 10

Fig. 1 Network topology containing nodes A and B

Fig. 2 Framework of DQN-based CPN resource matching method

Fig. 3 DQN structure

Fig. 4 Workflow of DQN-based CPN resource matching method

Fig. 5 Performance of loss function under different learning rates （lr） and batch sizes （bs）

Tab. 1 Return values under different discount factors

折扣因子	回报值	折扣因子	回报值	折扣因子	回报值
0.2	172.02	0.6	176.61	0.9	180.44
0.4	174.20	0.8	177.51

Fig. 6 Task request reception rates under different task numbers

Fig. 7 Average RAM utilizations under different task numbers

Fig. 8 Average BW utilizations under different task numbers

Fig. 9 Average STORAGE utilizations under different task numbers

References 33

[1]	PREMALATHA B， PRAKASAM P. Optimal energy-efficient resource allocation and fault tolerance scheme for task offloading in IoT-FoG computing networks［J］. Computer Networks， 2024， 238： No.110080.
[2]	SUN Y， LEI B， LIU J， et al. Computing power network： a survey［J］. China Communications， 2024， 21（9）： 109-145.
[3]	ZHANG D G， DONG W M， ZHANG T， et al. New computing tasks offloading method for MEC based on prospect theory framework［J］. IEEE Transactions on Computational Social Systems， 2024， 11（1）： 770-781.
[4]	OUYANG Y， YE X， SUN J， et al. The first decade of computing and network convergence［C］// Proceedings of the 2023 IEEE International Conference on Communications. Piscataway： IEEE， 2023： 1928-1933.
[5]	SHI X， LI Q， WANG D， et al. Mobile Computing Force Network （MCFN）： computing and network convergence supporting integrated communication service［C］// Proceedings of the 2022 International Conference on Service Science. Piscataway： IEEE， 2022： 131-136.
[6]	SONG L， HU X， ZHANG G， et al. Networking systems of AI： on the convergence of computing and communications［J］. IEEE Internet of Things Journal， 2022， 9（20）： 20352-20381.
[7]	TANG Q， XIE R， FENG L， et al. SIaTS： a service intent-aware task scheduling framework for computing power networks［J］. IEEE Network， 2023， 38（4）： 233-240.
[8]	GONG X， REN S， WANG C， et al. Research on computing resource measurement and routing methods in software defined computing first network［J］. Sensors， 2024， 24（4）： No.1086.
[9]	GONG X， BAI C， REN S， et al. Cloud-network resource perception and modeling technology based on software-defined computing first network［C］// Proceedings of the 3rd International Conference on Computer Science， Electronic Information Engineering and Intelligent Control Technology. Piscataway： IEEE， 2023： 6-12.
[10]	SUN W， LI Z， WANG Q， et al. FedTAR： task and resource-aware federated learning for wireless computing power networks［J］. IEEE Internet of Things Journal， 2023， 10（5）： 4257-4270.
[11]	LIN L， CHEN Y， ZHOU Z， et al. When Metaverse meets computing power networking： an energy-efficient framework for service placement［J］. IEEE Wireless Communications， 2023， 30（5）： 76-85.
[12]	LIU Y， WANG D， SONG B， et al. Green heterogeneous computing powers allocation using reinforcement learning in SDN-IoV［J］. IEEE Transactions on Green Communications and Networking， 2023， 7（2）： 983-995.
[13]	JIA Q， HU Y， ZHOU X， et al. Deterministic computing power networking： architecture， technologies and prospects［EB/OL］. ［2024-07-23］..
[14]	GUO Y， XU X， XIAO F. MADRLOM： a computation offloading mechanism for software-defined cloud-edge computing power network［J］. Computer Networks， 2024， 245： No.110352.
[15]	TANG S， YU Y， WANG H， et al. A survey on scheduling techniques in computing and network convergence［J］. IEEE Communications Surveys and Tutorials， 2023， 26（1）： 160-195.
[16]	SANG Y， WEI J， ZHANG Z， et al. A mobility-aware task scheduling by hybrid PSO and GA for mobile edge computing［J］. Cluster Computing， 2024： 1-16.
[17]	ZHANG Z， LI Q， LU L， et al. Joint optimization of the partition and scheduling of DNN tasks in computing and network convergence［J］. IEEE Networking Letters， 2023， 5（2）： 130-134.
[18]	LI J， LV H， LEI B， et al. Modeling and optimization for computing power resource-aware in CPN［C］// Proceedings of the 24th Asia-Pacific Network Operations and Management Symposium. Piscataway： IEEE， 2023： 353-356.
[19]	XING W， DONG G， WU N， et al. A URL-based computing power identification method［C］// Proceedings of the IEEE 5th International Conference on Electronics and Communication Engineering. Piscataway： IEEE， 2022： 99-103.
[20]	LI J， LV H， LEI B， et al. A computing power resource modeling approach for computing power network［C］// Proceedings of the 2022 International Conference on Computer Communications and Networks. Piscataway： IEEE， 2022： 1-2.
[21]	吴美希，杨晓彤.算力五力模型：一种衡量算力的综合方法［J］. 信息通信技术与政策，2022（3）：13-21.
	WU M X， YANG X T. Five forces of computational power： a comprehensive method to measure computational power［J］. Information and Communications Technology and Policy， 2022（3）： 13-21.
[22]	柴若楠，郜帅，兰江雨，等.算力网络中高效算力资源度量方法［J］.计算机研究与发展，2023，60（4）：763-771.
	CHAI R N， GAO S， LAN J Y， et al. Efficient computing resource metric method in computing-first network［J］. Journal of Computer Research and Development， 2023， 60（4）： 763-771.
[23]	夏天豪，夏长清，潘昊，等.基于强化学习的算力资源度量方法［J］.燕山大学学报，2023，47（3）：246-254.
	XIA T H， XIA C Q， PAN H， et al. Computational power resource measurement method based on reinforcement learning［J］. Journal of Yanshan University， 2023， 47（3）： 246-254.
[24]	LIU J， SUN Y， SU J， et al. Computing power network： a testbed and applications with edge intelligence［C］// Proceedings of the 2022 IEEE Conference on Computer Communications Workshops. Piscataway： IEEE， 2022： 1-2.
[25]	ZHANG J， GAO S， HOU X， et al. Resource awareness mechanism based on multi-dimensional identifier in compute first networking［C］// Proceedings of the 9th International Conference on Computer and Communication Systems. Piscataway： IEEE， 2024： 605-610.
[26]	LUAN Q， CUI H， YU X， et al. Task rescheduling for UAV-assisted emergency communications under uncertainty［C］// Proceedings of the 2023 IEEE Global Communications Conference. Piscataway： IEEE， 2023： 3039-3044.
[27]	JIANG H， WANG S， TANG C， et al. Reschedulable task allocation strategy in cloud-edge-end cooperative mobile crowd sensing［C］// Proceedings of the 2024 IEEE International Conference on Communications. Piscataway： IEEE， 2024： 3262-3267.
[28]	LI X， ZHOU Z， ZHAO Z， et al. Data & computation-intensive service re-scheduling in edge networks［C］// Proceedings of the 2021 IEEE International Conference on Web Services. Piscataway： IEEE， 2021： 389-396.
[29]	NAIR B， BHANU S M S. A reinforcement learning algorithm for rescheduling preempted tasks in fog nodes［J］. Journal of Scheduling， 2022， 25（5）： 547-565.
[30]	ZHUANG L， TIAN S， HE M， et al. Virtual network embedding algorithm via diffusion wavelet［J］. IEEE Access， 2019， 7： 134145-134157.
[31]	曾庆丰，蔡延光，胡城，等.蝗虫优化算法综述［J］.自动化与信息工程，2024，45（1）：1-11.
	ZENG Q F， CAI Y G， HU C， et al. A survey of grasshopper optimization algorithm［J］. Automation & Information Engineering， 2024，45（1）：1-11.
[32]	WU Z， LIU H， SHI X， et al. A resource scheduling strategy based on firefly algorithm［C］// Proceeding of the 2023 International Conference on Culture-Oriented Science and Technology. Piscataway： IEEE， 2023： 262-266.
[33]	SUKNUM S， THOASIRI C， JINAPORN N. Q-learning-based resource allocation in heterogeneous cellular networks［C］// Proceeding of the 2022 International Electrical Engineering Congress. Piscataway： IEEE， 2022： 1-3.

Highly reliable matching method based on multi-dimensional resource measurement and rescheduling in computing power network

算力网络中基于多维资源度量和重调度的高可靠匹配方法

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 10

References 33

Related Articles 12

Recommended Articles

Metrics

[1]	Tianyu XUE, Aiping LI, Liguo DUAN. Vehicular edge computing scheme with task offloading and resource optimization [J]. Journal of Computer Applications, 2025, 45(6): 1766-1775.
[2]	Huahua WANG, Liang HUANG, Jiajie CHEN, Jiening FANG. Dynamic allocation algorithm for multi-beam subcarriers of low orbit satellites based on deep reinforcement learning [J]. Journal of Computer Applications, 2025, 45(2): 571-577.
[3]	Xiaoyan ZHAO, Wei HAN, Junna ZHANG, Peiyan YUAN. Collaborative offloading strategy in internet of vehicles based on asynchronous deep reinforcement learning [J]. Journal of Computer Applications, 2024, 44(5): 1501-1510.
[4]	Jiachen YU, Ye YANG. Irregular object grasping by soft robotic arm based on clipped proximal policy optimization algorithm [J]. Journal of Computer Applications, 2024, 44(11): 3629-3638.
[5]	Tengfei CAO, Yanliang LIU, Xiaoying WANG. Edge computing and service offloading algorithm based on improved deep reinforcement learning [J]. Journal of Computer Applications, 2023, 43(5): 1543-1550.
[6]	Yidi WANG, Zhiwei LI, Wenxin ZHANG, Tieke LI, Bailin WANG. Estimation of distribution algorithm for hot rolling rescheduling with order disturbance [J]. Journal of Computer Applications, 2022, 42(8): 2628-2636.
[7]	Yu XU, Yunyou ZHU, Xiao LIU, Yuting DENG, Yong LIAO. Multi-objective routing optimization of electric power material distribution based on deep reinforcement learning [J]. Journal of Computer Applications, 2022, 42(10): 3252-3258.
[8]	WANG Yonghang, ZHANG Tianyu, ZHENG Hongxing. Multi-stage rescheduling method of liner considering severe weather [J]. Journal of Computer Applications, 2021, 41(1): 286-294.
[9]	WANG Zhongmin, WANG Ke, HE Yan. Multiple classifier fusion model for activity recognition based on high reliability weighted [J]. Journal of Computer Applications, 2016, 36(12): 3353-3357.
[10]	HOU Dong-Liang LI Tie-ke. Rescheduling algorithm for steelmaking and continuous casting based on dynamic constraint satisfaction [J]. Journal of Computer Applications, 2012, 32(12): 3553-3557.
[11]	. Data fusion weight μ-ξ of MP_WSMN to achieve high QoS [J]. Journal of Computer Applications, 2010, 30(06): 1469-1471.
[12]	. Design and implementation of RAID mechanism based on ReedSolomon algorithm [J]. Journal of Computer Applications, 2006, 26(11): 2756-2758.