CN112188632A - Ocean fog node online resource management method based on second-order gradient estimation - Google Patents

Ocean fog node online resource management method based on second-order gradient estimation Download PDF

Info

Publication number
CN112188632A
CN112188632A CN202011073234.0A CN202011073234A CN112188632A CN 112188632 A CN112188632 A CN 112188632A CN 202011073234 A CN202011073234 A CN 202011073234A CN 112188632 A CN112188632 A CN 112188632A
Authority
CN
China
Prior art keywords
resource allocation
fog node
computing
allocation period
resource
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
CN202011073234.0A
Other languages
Chinese (zh)
Inventor
徐艳丽
唐浩
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Maritime University
Original Assignee
Shanghai Maritime University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Maritime University filed Critical Shanghai Maritime University
Priority to CN202011073234.0A priority Critical patent/CN112188632A/en
Publication of CN112188632A publication Critical patent/CN112188632A/en
Withdrawn legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/50—Allocation or scheduling criteria for wireless resources
    • H04W72/53—Allocation or scheduling criteria for wireless resources based on regulatory allocation policies
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention provides an ocean fog node online resource management method based on second-order gradient estimation. The method comprises the following implementation steps: initializing data; the method comprises the steps that a fog node obtains a resource allocation action in a resource allocation period; the method comprises the steps that a fog node acquires a cost value corresponding to a resource allocation action in a resource allocation period; calculating gradient estimation parameters; and the fog node updates the computing resource allocation action for the next resource allocation period, and determines to allocate the computing resources of the fog node to the collected data in real time in the resource allocation period according to the computing resource allocation action. Under the condition that the constraint function is known, aiming at the situation that maritime communication resources are limited, computing resources at fog nodes are scheduled for different application data on line, and data are preprocessed by using the distributed computing resources, so that the communication resources for transmitting the data are saved, the communication service quality of different applications is ensured, meanwhile, the transmission efficiency of the communication resources is effectively improved, and the maritime communication cost is reduced.

Description

Ocean fog node online resource management method based on second-order gradient estimation
Technical Field
The invention relates to the technical field of communication, in particular to an ocean fog node online resource management method based on second-order gradient estimation.
Background
With the increasing frequency of human marine activities, various devices for marine activities are increasing, which generate a large amount of data such as marine environment observation, marine organism activity monitoring, weather statistics, etc., which are received, processed and transmitted by the fog nodes on the ships and buoys, and the base stations and the satellites receive the data sent by the fog nodes and transmit the data to shore-based users, so as to achieve the purpose of uniform management, analysis and application of marine data. Unlike terrestrial communication, maritime communication has the problems of limited communication resources, lack of infrastructure communication facilities, high communication cost and the like, and the most urgent problem to be solved is how to efficiently utilize the limited communication resources to transmit data to improve maritime communication efficiency. The current resource management method does not effectively solve the problem, and does not consider real-time data distribution at the fog node by using the computing resource of the fog node, so that the data is preprocessed by using the distributed computing resource to save communication resources for transmitting the data, improve the maritime communication efficiency, and meet the service quality requirements of different marine applications.
Disclosure of Invention
The invention provides a marine fog node online resource management method based on second-order gradient estimation, which does not need display expression of a cost function and a constraint function, and distributes data in real time by reasonably scheduling computing resources at fog nodes under the condition that the constraint function is known, so that the data is preprocessed by utilizing the distributed computing resources to save communication resources for transmitting the data, improve the marine communication efficiency and meet the service quality requirements of different marine applications.
The invention is realized by the following technical scheme:
a marine fog node online resource management method based on second-order gradient estimation is characterized by comprising the following steps:
s1, pair
Figure BDA0002715835270000011
Initializing, t is 1, wherein xtIndicating the resource allocation actions of the fog node in the t-th resource allocation period, wherein the resource allocation actions comprise the computing resources which can be allocated to the data by the fog node in the resource allocation period, the computing resources are virtualized,without units, the real-world can be computational power, can be memory size, function of frequency, and ΛtRepresenting the gradient parameter at the t-th resource allocation period,
Figure BDA0002715835270000012
Figure BDA0002715835270000013
c represents a formula
Figure BDA0002715835270000014
Boundary of (a)tExpressing the dual variable at the t-th resource allocation period by the formula
Figure BDA0002715835270000015
Updating dual variable lambda for next resource allocation periodt+1，[x]+Max { x,0}, d denotes a dimension, denotes a preselected constant greater than 0, α and μ denote predefined constants, β denotes a constant of positive step size, u denotes a constant of positive step sizen,tRepresents a surface independent of the unit ball;
s2, fog node is based on formula
Figure BDA0002715835270000021
Acquiring resource allocation action x in the t-th resource allocation periodtAnd xt+un,tWherein N is 2 … N,
Figure BDA0002715835270000022
Figure BDA0002715835270000023
x represents the feasible set of settings, since the cost function is unknown in the settings, the gradient of the cost is represented as:
Figure BDA0002715835270000024
looping from step S2 to step S5, and adding 1 to t once per loop;
s3, the fog node obtains the resource allocation action in the t resource allocation periodxtAnd xt+un,tCorresponding cost value ft(xt)Hz/bit，ft(xt+un,t)Hz/bit；
S4, calculating gradient estimation parameters
Figure BDA0002715835270000025
And
Figure BDA0002715835270000026
s5, fog node is based on formula
Figure BDA0002715835270000027
Updating the next resource allocation period update computing resource allocation action Pt+1Allocating actions P according to computing resourcestAnd deciding to distribute the computing resources of the fog nodes to the collected data in real time in a resource distribution period.
The invention has the beneficial effects that: the method does not need display expression of a cost function and a constraint function, and under the condition that the constraint function is known, the method allocates data in real time by reasonably scheduling the computing resources at the fog nodes, so that the data is preprocessed by using the allocated computing resources, the communication resources for transmitting the data are saved, the maritime communication efficiency is improved, and the service quality requirements of different marine applications are met.
Drawings
FIG. 1 is a flow chart of an ocean fog node online resource management method based on second-order gradient estimation according to the present invention;
FIG. 2 is a comparison of average cost variation of different resource management schemes of the ocean fog node online resource management method based on second-order gradient estimation under different computing resource quantities.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments.
The invention provides an ocean fog node online resource management method based on second-order gradient estimation. Fig. 1 shows a flow chart of an ocean fog node online resource management method based on second-order gradient estimation, which is characterized by comprising the following steps:
s1, pair
Figure BDA0002715835270000031
Initializing, t is 1, wherein xtThe resource allocation action of the fog node in the t-th resource allocation period is represented, wherein the resource allocation action comprises a computing resource which can be allocated to data by the fog node in the resource allocation period, the computing resource is virtualized and has no unit, and the computing resource can be computing power in reality and can be a function of the size and frequency of a memory, and the lambdatRepresenting the gradient parameter at the t-th resource allocation period,
Figure BDA0002715835270000032
Figure BDA0002715835270000033
c represents a formula
Figure BDA0002715835270000034
Boundary of (a)tExpressing the dual variable at the t-th resource allocation period by the formula
Figure BDA0002715835270000035
Updating dual variable lambda for next resource allocation periodt+1，[x]+Max { x,0}, d denotes a dimension, denotes a preselected constant greater than 0, α and μ denote predefined constants, β denotes a constant of positive step size, u denotes a constant of positive step sizen,tRepresents a surface independent of the unit ball;
s2, fog node is based on formula
Figure BDA0002715835270000036
Acquiring resource allocation action x in the t-th resource allocation periodtAnd xt+un,tWherein N is 2 … N,
Figure BDA0002715835270000037
Figure BDA0002715835270000038
x represents the feasible set of settings, since the cost function is unknown in the settings, the gradient of the cost is represented as:
Figure BDA0002715835270000039
looping from step S2 to step S5, and adding 1 to t once per loop;
s3, the fog node acquires the resource allocation action x in the t-th resource allocation periodtAnd xt+un,tCorresponding cost value ft(xt)Hz/bit，ft(xt+un,t)Hz/bit；
S4, calculating gradient estimation parameters
Figure BDA00027158352700000310
And
Figure BDA00027158352700000311
s5, fog node is based on formula
Figure BDA0002715835270000041
Updating the next resource allocation period update computing resource allocation action Pt+1Allocating actions P according to computing resourcestAnd deciding to distribute the computing resources of the fog nodes to the collected data in real time in a resource distribution period.
The beneficial effects of the invention are further verified by corresponding experimental data as follows:
by picking five application data with different delay constraints and operating them at the fog node. Since the marine application data generation rate is low and the marine data transmission rate is of the kbps level, the data arrival rate of the application i is set to γi1000 × i. The observation time is selected to be T equal to 2000ms, and the step length is selected to be mu equal to alpha equal to 0.05/T. Data such as observation time, resources, and computing power of the cloud node are only used for the experiment, and the data can be adjusted accordingly for different situations.
Fig. 2 shows a comparison of average cost variation for different resource management schemes at different amounts of computing resources. The horizontal axis is the number of computing resources used for preprocessing data in different resource management schemes, the vertical axis is the average cost Hz/bit under the number of the computing resources, and in observation time, the average cost is continuously reduced on the contrary along with the increase of the number of the computing resources in the proposed scheme and the unified scheme. The average cost changes irregularly for the random scheme because the random scheme randomly selects the computing resources and does not depend on the change of the number of the computing resources. It is clear that the average cost of the proposed solution compared to the other two remains at a minimum. The proposed scheme determines that the next resource allocation period schedules resources in the most reasonable way by observing the resource condition of the previous resource allocation period.
The above description is only an example of the present invention, and the scope of the present invention is not limited thereto, and it should be apparent to those skilled in the art that various equivalent modifications or substitutions can be made within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (1)

1. A marine fog node online resource management method based on second-order gradient estimation is characterized by comprising the following steps:
s1, for x1＝0,Λ0＝∈Id,
Figure FDA0002715835260000011
Initializing, t is 1, wherein xtThe resource allocation action of the fog node in the t-th resource allocation period is represented, wherein the resource allocation action comprises a computing resource which can be allocated to data by the fog node in the resource allocation period, the computing resource is virtualized and has no unit, and the computing resource can be computing power in reality and can be a function of the size and frequency of a memory, and the lambdatRepresenting the gradient parameter at the t-th resource allocation period,
Figure FDA0002715835260000012
Figure FDA0002715835260000013
c represents a formula
Figure FDA0002715835260000014
Boundary of (a)tExpressing the dual variable at the t-th resource allocation period by the formula
Figure FDA0002715835260000015
Updating dual variable lambda for next resource allocation periodt+1，[x]+Max { x,0}, d denotes a dimension, denotes a preselected constant greater than 0, α and μ denote predefined constants, β denotes a constant of positive step size, u denotes a constant of positive step sizen,tRepresents a surface independent of the unit ball;
s2, fog node is based on formula
Figure FDA0002715835260000016
Acquiring resource allocation action x in the t-th resource allocation periodtAnd xt+un,tWherein N is 2 … N,
Figure FDA0002715835260000017
Figure FDA0002715835260000018
x represents the feasible set of settings, since the cost function is unknown in the settings, the gradient of the cost is represented as:
Figure FDA0002715835260000019
looping from step S2 to step S5, and adding 1 to t once per loop;
s3, the fog node acquires the resource allocation action x in the t-th resource allocation periodtAnd xt+un,tCorresponding cost value ft(xt)Hz/bit，ft(xt+un,t)Hz/bit；
S4, calculating gradient estimation parameters
Figure FDA00027158352600000110
And
Figure FDA00027158352600000111
s5, fog node is based on formula
Figure FDA00027158352600000112
Updating the next resource allocation period update computing resource allocation action Pt+1Allocating actions P according to computing resourcestAnd deciding to distribute the computing resources of the fog nodes to the collected data in real time in a resource distribution period.
CN202011073234.0A 2020-10-09 2020-10-09 Ocean fog node online resource management method based on second-order gradient estimation Withdrawn CN112188632A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011073234.0A CN112188632A (en) 2020-10-09 2020-10-09 Ocean fog node online resource management method based on second-order gradient estimation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011073234.0A CN112188632A (en) 2020-10-09 2020-10-09 Ocean fog node online resource management method based on second-order gradient estimation

Publications (1)

Publication Number Publication Date
CN112188632A true CN112188632A (en) 2021-01-05

Family

ID=73948702

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011073234.0A Withdrawn CN112188632A (en) 2020-10-09 2020-10-09 Ocean fog node online resource management method based on second-order gradient estimation

Country Status (1)

Country Link
CN (1) CN112188632A (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014196716A1 (en) * 2013-06-03 2014-12-11 엘지전자 주식회사 Method for managing wireless resource and apparatus therefor
CN109189563A (en) * 2018-07-25 2019-01-11 腾讯科技（深圳）有限公司 Resource regulating method, calculates equipment and storage medium at device
CN109951869A (en) * 2019-03-01 2019-06-28 重庆邮电大学 A kind of car networking resource allocation methods calculated based on cloud and mist mixing
CN110087318A (en) * 2019-04-24 2019-08-02 重庆邮电大学 Task unloading and resource allocation joint optimization method based on the mobile edge calculations of 5G
CN110519776A (en) * 2019-08-07 2019-11-29 东南大学 Balanced cluster and federated resource distribution method in a kind of mist computing system
EP3709708A1 (en) * 2019-03-15 2020-09-16 Nokia Solutions and Networks Oy Method and apparatus for configuring a communication network

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014196716A1 (en) * 2013-06-03 2014-12-11 엘지전자 주식회사 Method for managing wireless resource and apparatus therefor
CN109189563A (en) * 2018-07-25 2019-01-11 腾讯科技（深圳）有限公司 Resource regulating method, calculates equipment and storage medium at device
CN109951869A (en) * 2019-03-01 2019-06-28 重庆邮电大学 A kind of car networking resource allocation methods calculated based on cloud and mist mixing
EP3709708A1 (en) * 2019-03-15 2020-09-16 Nokia Solutions and Networks Oy Method and apparatus for configuring a communication network
CN110087318A (en) * 2019-04-24 2019-08-02 重庆邮电大学 Task unloading and resource allocation joint optimization method based on the mobile edge calculations of 5G
CN110519776A (en) * 2019-08-07 2019-11-29 东南大学 Balanced cluster and federated resource distribution method in a kind of mist computing system

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
JIACHEN YANG: "Fog-based marine environmental information monitoring toward ocean of things", 《IEEE INTERNET OF THINGS JOURNAL》 *
YANLI XU: "Quality of service provisions for maritime communications based on cellular networks", 《IEEE ACCESS》 *
乔冠华: "基于移动边缘计算的物联网资源管理策略研究", 《中国优秀博士论文电子期刊网》 *
谢人超: "认知无线网络资源管理算法研究", 《中国优秀博士论文电子期刊网》 *

Similar Documents

Publication Publication Date Title
CN109829332B (en) Joint calculation unloading method and device based on energy collection technology
He et al. Dynamic scheduling of hybrid tasks with time windows in data relay satellite networks
CN107392382B (en) High-resolution geostationary orbit imaging satellite observation task planning method
Liu et al. Task scheduling in fog enabled Internet of Things for smart cities
CN111629380A (en) Dynamic resource allocation method for high-concurrency multi-service industrial 5G network
CN113469325A (en) Layered federated learning method, computer equipment and storage medium for edge aggregation interval adaptive control
CN114885420A (en) User grouping and resource allocation method and device in NOMA-MEC system
CN112559147B (en) Dynamic matching method, system and equipment based on GPU (graphics processing Unit) occupied resource characteristics
CN113407249A (en) Task unloading method facing to position privacy protection
CN114116061B (en) Workflow task unloading method and system in mobile edge computing environment
CN112188632A (en) Ocean fog node online resource management method based on second-order gradient estimation
Xu et al. Cached model-as-a-resource: Provisioning large language model agents for edge intelligence in space-air-ground integrated networks
Yang et al. Edge computing and networking resource management for decomposable deep learning: An auction-based approach
Chen et al. Latency-optimal pyramid-based joint communication and computation scheduling for distributed edge computing
CN116302569B (en) Resource partition intelligent scheduling method based on user request information
CN112312299A (en) Service unloading method, device and system
CN114979027B (en) Offshore edge computing network task unloading and resource allocation method based on energy collection
CN110392377B (en) 5G ultra-dense networking resource allocation method and device
CN112188631A (en) Ocean fog node resource management method based on gradient estimation
CN116843016A (en) Federal learning method, system and medium based on reinforcement learning under mobile edge computing network
CN115543582A (en) Method, system and equipment for unified scheduling of super computing power network
CN113518112B (en) Wireless charging equipment cooperative task unloading strategy based on multi-agent reinforcement learning
CN115499876A (en) Computing unloading strategy based on DQN algorithm under MSDE scene
CN115733838A (en) Vehicle networking multidimensional resource allocation method based on mobile edge calculation
Khaytbaev Decision routing problems in a wireless sensor network based on a neural mechanism

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
WW01 Invention patent application withdrawn after publication

Application publication date: 20210105

WW01 Invention patent application withdrawn after publication