WO2024222020A1 - 一种随机性在线云边端协同的数据存储方法及系统 - Google Patents

一种随机性在线云边端协同的数据存储方法及系统 Download PDF

Info

Publication number
WO2024222020A1
WO2024222020A1 PCT/CN2023/142749 CN2023142749W WO2024222020A1 WO 2024222020 A1 WO2024222020 A1 WO 2024222020A1 CN 2023142749 W CN2023142749 W CN 2023142749W WO 2024222020 A1 WO2024222020 A1 WO 2024222020A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
edge
cloud
storage
random
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.)
Pending
Application number
PCT/CN2023/142749
Other languages
English (en)
French (fr)
Inventor
谢永杰
潘晓东
李伟泽
黎达伟
罗熙
赵学慧
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.)
China Telecom Cloud Technology Co Ltd
Original Assignee
China Telecom Cloud Technology Co Ltd
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 China Telecom Cloud Technology Co Ltd filed Critical China Telecom Cloud Technology Co Ltd
Publication of WO2024222020A1 publication Critical patent/WO2024222020A1/zh
Anticipated expiration legal-status Critical
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0668Interfaces specially adapted for storage systems adopting a particular infrastructure
    • G06F3/067Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10Complex mathematical operations
    • G06F17/18Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0628Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0629Configuration or reconfiguration of storage systems
    • G06F3/0631Configuration or reconfiguration of storage systems by allocating resources to storage systems
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601Interfaces specially adapted for storage systems
    • G06F3/0628Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0646Horizontal data movement in storage systems, i.e. moving data in between storage devices or systems
    • G06F3/0647Migration mechanisms

Definitions

  • the present invention belongs to the field of cloud-edge-end collaboration and cloud storage technology, and in particular to a random online cloud-edge-end collaborative data storage method and system.
  • cloud computing has maintained a rapid development momentum and is also one of the hot topics in the field of IT technology development.
  • cloud computing can not only effectively reduce the cost of services, but also greatly improve the utilization of resources.
  • cloud computing also has many advantages such as elastic expansion, disaster recovery, and on-demand acquisition.
  • cloud computing continues to grow and develop in the fields of IaaS, PaaS, and SaaS, the surge in data volume has brought unprecedented storage pressure to major cloud service providers.
  • cloud storage systems with low storage costs and more intelligent systems. Thanks to the high efficiency, low cost, intelligence, and high reliability of cloud storage, cloud storage has been widely used in various production environments, such as general storage, big data analysis, database services, microservices, and other fields, and plays a key role.
  • the traditional cloud computing paradigm will face huge challenges in terms of data storage, workload, and bandwidth resources, which will lead to many problems in application scenarios with low latency and high bandwidth requirements, such as Internet of Vehicles and intelligent monitoring.
  • data stored in the cloud through cloud-edge-end collaboration, data that requires real-time calculation and analysis can be more efficiently stored close to the terminal device to ensure the real-time data processing.
  • the "cloud-edge-end” architecture the "cloud” is the central node, responsible for the management and control of edge computing
  • the “edge” is the edge side of cloud computing, responsible for local data analysis
  • the “end” is the terminal device, responsible for data collection, perception and other operations.
  • the cloud-edge-end collaboration mechanism can not only effectively reduce the storage costs of cloud users, but also provide cloud users with Better quality service.
  • storing data in the cloud often saves more costs compared to setting up a large number of storage servers at the edge for data storage. If some data is frequently accessed on the edge over a period of time, it is obviously more appropriate to store this data on the edge. On the one hand, it can ensure the real-time nature of the data during transmission as much as possible, and on the other hand, it can effectively reduce the bandwidth overhead of the data during transmission. On the contrary, if some data is accessed very rarely on the edge over a period of time, it is obviously more appropriate to store this data in the cloud, which can not only effectively reduce the cost of data storage, but also effectively alleviate the pressure of data storage on the edge.
  • patent CN111800486A provides a cloud-edge collaborative resource scheduling method and system.
  • the invention makes reservation decisions in real time according to the actual bandwidth occupancy of users in various regions, thereby effectively reducing the surge in bandwidth costs caused by the surge in the number of visitors in a short period of time.
  • the invention monitors the access popularity of each resource by customers within the service range of each MEC server in real time to determine whether the access bandwidth of a certain resource customer in a certain region reaches the set threshold, and then determines whether the MEC server should be rented in the region or the MEC server and the cloud server should be used collaboratively to provide access during the reservation period of the rented MEC server.
  • the invention does not fully consider the QoS of users using cloud storage services, and its application scenarios have certain limitations.
  • the cloud-edge collaborative resource scheduling method proposed by the invention is relatively simple, and there is a lot of room for optimization in algorithm design to save more storage costs.
  • Patent CN113157446A provides a cloud-edge collaborative resource allocation method, device, equipment and medium.
  • the invention controls resources through blockchain and cloud-edge collaborative solutions, starts all subtasks according to the coupling rule logic, and finally completes the total task according to the aggregation of all subtask execution results.
  • the tasks that need to allocate resources are analyzed through the cloud-edge collaborative platform to determine the task parameters and the minimum split subset of the resources to be allocated, and the cloud-edge collaborative solution is determined based on the task parameters of the resources to be allocated and the task resource collaborative algorithm, and finally the resources are allocated to the tasks of the resources to be allocated according to the blockchain and the cloud-edge collaborative solution.
  • the invention introduces blockchain technology, which will cause a large amount of energy consumption, and with the surge in data volume, performance problems of varying degrees will be derived.
  • the invention does not consider reducing the cost of data storage, but only proposes a single idealized cloud-edge collaborative resource allocation strategy.
  • cloud service providers have paid more and more attention to optimizing cloud costs, controlling cloud waste and improving cloud deployment efficiency.
  • the edge side has a stronger demand for data perception. Therefore, how to fully reduce the cost of cloud storage and improve the deployment efficiency of cloud storage services from the perspective of cloud-edge-end collaboration has important research significance and value.
  • the current cloud-edge collaborative resource scheduling mechanism can no longer meet the task processing requirements of current application scenarios.
  • the purpose of the present invention is to provide a random online cloud-edge-end collaborative data storage method and system to address the deficiencies in the prior art. It provides an intelligent, flexible and efficient data storage strategy in real time based on the historical storage behavior of the data and the frequency of data processing on the cloud and edge sides. While ensuring data storage performance, it greatly reduces the deployment cost of cloud storage services and improves the deployment efficiency of cloud storage services.
  • An embodiment of the present application provides a random online cloud-edge-device collaborative data storage method, the method comprising:
  • Step 1 Start selecting the research object. Randomly select an availability zone as the object within a geographical area and count the number n of edge nodes in the availability zone AZ.
  • Step 2 Set relevant parameters, let b center be the amount of data stored in the cloud, let is the amount of data stored at the edge, p center is the unit price of data storage in the cloud, and is the unit price of data storage at the edge, where i ⁇ [1,n], let q be the unit price of data access, let t be the storage duration of data, and let T be the storage billing cycle. Assuming that the data is stored in the cloud in the past storage cycle, let its storage cost be C center , assuming that the data is stored at the edge in the past storage cycle, let its storage cost be C edge .
  • Step 3 Set the size of the sliding window and count related data.
  • Set the size of the sliding window z monitor the data requests of the terminal devices in real time, record and count the number of data access requests m initiated by cloud users at the terminal within the sliding window range, and count the number of times these data are processed in the cloud and the edge.
  • the number of times data is processed in the cloud is recorded as h center
  • the number of times data is processed at the edge is recorded as h edge .
  • Step 4 Calculate the elastic equilibrium point ⁇ e , construct the probability density function, and obtain the random factor k;
  • Step 5 Design a random online cloud-edge collaborative data storage algorithm
  • Step 6 Determine whether the current data needs to be migrated through the machine-based online cloud-edge collaborative data storage algorithm.
  • the random online cloud-edge-device collaborative data storage method further includes:
  • the data stored in the edge will be migrated to the cloud, or the data in the cloud will be migrated to the edge.
  • the random online cloud-edge collaborative data storage method is executed, and in the next storage period T', the method is repeated. If the current data does not need to be migrated, the process is terminated directly.
  • the calculation of the elastic balance point ⁇ e is a preliminary preparation for proposing a random online cloud-edge collaborative data storage algorithm.
  • the elastic balance point ⁇ e is obtained through the relevant parameters set in steps 1-3.
  • the probability density function is constructed by combining the calculation formula of the data storage cost and the set related parameters with the elastic equilibrium point ⁇ e obtained by calculation, where
  • the constructed probability density function is as follows:
  • ⁇ ( ⁇ ) is the Dirac delta function
  • k is a random factor
  • the constructed probability density function obtains the random factor k by calculating its inverse function, where Mathematical expectation of function f(y) Calculate the inverse function of f(k) to get the random factor k.
  • the calculation formula of the random factor k is:
  • step 5 the design of a random online cloud-edge collaborative data storage algorithm includes the following steps:
  • the size of the random factor k determines whether the data should be stored in the cloud or the edge. For data that needs to be physically migrated, it is first stored in the queue n i for caching based on the principle of locality, and then the data migration is completed.
  • Another embodiment of the present application provides a random online cloud-edge-end collaborative data storage system, the system comprising:
  • the data collection module obtains resource request information from the cloud, edge, and terminals, and receives data access record information connected to the server;
  • Real-time monitoring module which monitors data access requests of the cloud, edge and terminal in real time, including parameters such as request time and data volume;
  • Data storage module which stores collected data, set parameters and real-time monitoring information
  • Calculation module calculates elastic equilibrium point, constructs probability density function, obtains random factors, calculates inverse function, calculates mathematical expectation and other key data;
  • the algorithm generation module generates a random online cloud-edge collaborative data storage algorithm to determine whether the current data should be stored in the cloud or the edge.
  • the collaborative scheduling module migrates data stored in the cloud to the edge, or migrates data stored in the edge to the cloud;
  • the log module records information such as terminal user resource requests and data acquisition, and records the migration of data between the cloud and edge.
  • Another embodiment of the present application provides a storage medium for storing computer instructions, which, when executed by a processor, completes a random online cloud-edge-end collaborative data storage method.
  • a terminal device including a memory, a processor and a computer program, wherein the memory stores the computer program, and the processor is configured to run the computer program to implement any of the methods described above.
  • the present invention can make reasonable decisions in real time on whether data should be stored at the edge or in the cloud, and reduce the deployment cost of cloud storage services as much as possible, improve the deployment efficiency of cloud storage services, and at the same time ensure the QoS of users when using cloud storage services.
  • a random online cloud-edge collaborative data storage algorithm is proposed. By combining the random online cloud-edge collaborative data storage algorithm with probability distribution and mathematical expectation, the deployment cost of cloud storage services can be reduced as much as possible.
  • the random factors generated by the random online cloud-edge collaborative data storage algorithm can make more reasonable decisions in real time on whether data should be stored at the edge or in the cloud.
  • the concept of "sliding window” is introduced to realize real-time monitoring of data stored in the cloud and edge, and supports custom sliding window size, which can be adaptive The perception range of the cloud and edge for the stored data is adjusted accordingly.
  • the accuracy and flexibility of the prediction results of the random online cloud-edge collaborative data storage algorithm can be effectively improved, and it can easily cope with different production environment requirements and improve the deployment efficiency of cloud storage services.
  • the concept of "elastic-break-even-points” is proposed. Usually, the balance point is a constant calculated based on a certain equation, which can simplify the research model.
  • a multi-level queue coordination mechanism is introduced. By setting a multi-level queue, priority caching of data stored in the cloud and edge is achieved based on the principle of locality, so that the data access and use process is more stable, so as to fully guarantee the QoS of users when using cloud storage services.
  • FIG1 is a schematic diagram of a random online cloud-edge-device collaborative data storage method provided in Embodiment 1 of the present invention.
  • FIG2 is a schematic diagram of a random online cloud-edge collaborative data storage algorithm flow provided in Embodiment 1 of the present invention.
  • Figure 3 is a structural diagram of the architecture of a random online cloud-edge-end collaborative data storage system provided in Example 2 of the present invention.
  • the embodiment of the present invention first provides an editing information error correction method based on a rich text editor, which can be applied to terminal devices, such as computer terminals, specifically ordinary computers, quantum computers, etc.
  • Embodiment 1 is a diagrammatic representation of Embodiment 1:
  • FIG1 is a random online cloud-edge-device collaborative data storage method provided by an embodiment of the present invention, comprising the following steps:
  • Step 1 Select the research object. In a region, randomly select an availability zone as the object and count the number of edge nodes n in the availability zone AZ.
  • Step 2 Set relevant parameters. Let b center be the amount of data stored in the cloud, let is the amount of data stored at the edge, p center is the unit price of data storage in the cloud, and is the unit price of data storage at the edge, where i ⁇ [1,n], let q be the unit price of data access, let t be the storage duration of data, and let T be the storage billing cycle. Assuming that the data is stored in the cloud in the past storage cycle, let its storage cost be C center , assuming that the data is stored at the edge in the past storage cycle, let its storage cost be C edge .
  • Step 3 Set the size of the sliding window and count the relevant data.
  • Set the size of the sliding window z and monitor the data requests of the terminal device in real time. Record and count the number of data access requests m initiated by cloud users at the terminal within the sliding window. Count the number of times these data are processed in the cloud and the edge. The number of times data is processed in the cloud is recorded as h center , and the number of times data is processed at the edge is recorded as h edge .
  • Step 4 Perform mathematical operations through probability theory and inverse functions to calculate the elastic equilibrium point ⁇ e , construct a probability density function, and obtain the random factor k, so as to prepare for the subsequent random online cloud-edge collaborative data storage algorithm.
  • ⁇ ( ⁇ ) is the Dirac delta function
  • k is a random factor
  • Step 5 Random online cloud-edge collaborative data storage algorithm, whose execution process is shown in Figure 2, includes the following steps:
  • Step 6 In the current storage period T, the random online cloud-edge-end collaborative data storage method is executed. In the next storage period T', the method is repeated.
  • This embodiment describes in detail an implementation method of a random online tiered storage method for hot and cold data in the cloud.
  • users should not be limited to the method described in this embodiment and can make appropriate adjustments based on their own business and actual conditions.
  • Embodiment 2 is a diagrammatic representation of Embodiment 1:
  • FIG. 3 is a random online cloud-edge-end collaborative data storage system provided by an embodiment of the present invention.
  • the system includes seven modules, namely: data acquisition module, real-time monitoring module, data storage module, calculation module, algorithm generation module, collaborative scheduling module and log module.
  • the data acquisition module can obtain resource request information from the cloud, edge, and terminal, and receive data access record information connected to the server;
  • the real-time monitoring module can monitor data access requests from the cloud, edge, and terminal in real time, including parameters such as request time and data size;
  • the data storage module is responsible for storing the collected data, set parameters, and real-time monitoring information;
  • the calculation module is mainly used to calculate elastic break-even-point, construct probability density function, obtain random factors, calculate inverse function, calculate mathematical expectation and other key data, to provide support for generating random online cloud-edge collaborative data storage algorithms;
  • the algorithm generation module generates random online cloud-edge collaborative The same data storage algorithm determines whether the current data should be stored in the cloud or the edge.
  • the collaborative scheduling module is responsible for migrating data stored in the cloud to the edge, or migrating data stored on the edge to the cloud.
  • the log module is used to record information such as terminal user resource requests and data acquisition, and to record the migration of data on both the cloud and edge sides.
  • Embodiment 3 is a diagrammatic representation of Embodiment 3
  • This embodiment also provides an electronic device, including: one or more processors, one or more memories, and one or more computer programs; wherein the processor is connected to the memory, and the one or more computer programs are stored in the memory.
  • the processor executes the one or more computer programs stored in the memory so that the electronic device executes the method described in the above embodiment one.
  • the processor may be a central processing unit CPU, or other general-purpose processors, digital signal processors DSP, application-specific integrated circuits ASIC, off-the-shelf programmable gate arrays FPGA or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.
  • the memory may include a read-only memory and a random access memory, and provide instructions and data to the processor, and a portion of the memory may also include a non-volatile random access memory.
  • the memory may also store information about the device type.
  • each step of the method of Embodiment 1 can be completed by an integrated logic circuit of hardware in a processor or by instructions in the form of software.
  • a computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by a processor, the method described in embodiment 1 is completed.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Algebra (AREA)
  • Probability & Statistics with Applications (AREA)
  • Evolutionary Biology (AREA)
  • Operations Research (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

本发明公开了一种随机性在线云边端协同的数据存储方法及系统,方法包括:首先选取研究对象,在一个地域范围内,随机选取一个可用区作为对象,统计该可用区AZ中边缘端节点的数量n,继而设定相关参数,随后设定滑动窗口尺寸大小并统计相关数据,设定滑动窗口的尺寸大小z,实时监测终端设备的数据请求情况,再计算弹性平衡点βe,构造概率密度函数,获取随机因子k,最后设计随机性在线云边端协同数据存储算法,通过该算法判断当前数据是否需要进行迁移,它通过提供一种智能、灵活、高效的数据存储策略,在保证数据存储性能的同时极大地降低云存储服务的部署成本,提高云存储服务的部署效率。

Description

一种随机性在线云边端协同的数据存储方法及系统 技术领域
本发明属于云边端协同、云存储技术领域,特别是一种随机性在线云边端协同的数据存储方法及系统。
背景技术
近年来,伴随着数字经济的飞速发展,云计算一直保持着迅猛的发展势头,同时也是当今IT技术发展领域的热点话题之一。云计算作为计算技术和生产实践长期演变的产物,不仅可以有效降低服务的成本,也可以大幅提升资源的利用率。此外,云计算还具备着弹性扩展、容灾恢复、按需获取等诸多优点。
随着云计算在IaaS、PaaS、SaaS领域不断深耕并逐步发展壮大,数据量的激增给各大云服务提供商带来了前所未有的存储压力。相比于自建存储服务器所造成的硬件资源成本高、维护开销大等缺点,更多的互联网企业选择了存储成本低、更加智能化的云存储系统。得益于云存储的高效率、低成本、智能化、高可靠等特点,云存储被广泛地应用到各生产环境中,如:通用存储、大数据分析、数据库服务、微服务等领域,并发挥着关键的作用。
现如今,随着智能设备越来越多的接入网络,传统云计算范式中的数据存储、工作负载以及带宽资源等方面将面临巨大挑战,使得在车联网、智能监控等低延迟、高带宽需求的应用场景面临着诸多问题。对于存储在云端的数据,通过云边端协同可以更加高效的将需要实时计算和分析的数据存放到靠近终端设备的地方,以保证数据处理的实时性。在“云-边-端”架构中,“云”是中心节点,负责对边缘计算的管控,“边”是云计算的边缘侧,负责局部的数据分析,“端”是终端设备,负责数据的采集、感知等操作。云边端协同的机制,不仅可以有效降低云用户的存储成本,还可以在时空的维度为云用户提供 更加优质的服务。
对于云服务提供商而言,相较于在边缘端设立大量的存储服务器来进行数据存储,在云端存储数据往往会节省更多的成本。在一段时间内,如果某些数据在边缘侧被频繁访问,显然这些数据存放到边缘端更加合适,一方面可以尽可能地保证数据在传输过程中的实时性,另一方面可以有效降低数据在传输过程中带宽的开销。相反,如果某些数据一段时间内在边缘侧被访问的频次很低,显然这些数据存储在云端更加合适,这样不仅可以有效降低数据存储的成本,也可以有效缓解边缘侧数据存储的压力。
在现有的云边协同方案中,专利CN111800486A提供了一种云边协同的资源调度方法及系统。该发明在执行的过程中根据用户各地区带宽的实际占用情况实时地做出预留决策,从而有效减少由于访问人数短时间内的激增而导致的带宽费用的激增。该发明从访问带宽出发,通过实时监测各MEC服务器服务范围内客户对每一个资源的访问热度,来判断某地区某资源客户的访问带宽达到所设定的阈值,进而判断在该地区应当租用MEC服务器还是在所述租用MEC服务器的预留期内,协同使用MEC服务器和云服务器提供访问。然而,该发明没有充分考虑用户在使用云存储服务的QoS,其应用场景具有一定的局限性。同时,该发明提出的云边协同资源调度方法较为简单,在算法设计上有着很大的优化空间,以节约更多的存储成本。
专利CN113157446A提供了一种云边协同的资源分配方法、装置、设备及介质。该发明通过区块链和云边协同方案对资源进行限定控制,根据耦合规则逻辑启动所有子任务,最终根据所有的子任务执行成果聚合完成总任务。首先通过云边协同平台对需要分配资源的任务进行分析,确定所述待分配资源的任务参数以及最小拆分子集,并根据所述待分配资源的任务参数结合任务资源协同算法,确定云边协同方案,最终根据所述区块链以及所述云边协同方案,对所述待分配资源的任务进行资源分配。然而,该发明引入了区块链技术,会造成能源的大量消耗,伴随着数据量的激增,会衍生出不同程度的性能问题。此外,该发明没有考虑降低数据存储的成本,仅是单一的提出了一种理想化的云边协同资源分配策略。
近年来云服务提供商日益重视优化云成本、控制云浪费和提高云部署效率,加之越来越多的智能设备接入网络,边缘侧对数据的感知需求愈发强烈,因此如何从云边端协同的角度来充分降低云存储的成本,提高云存储服务的部署效率,具有重要的研究意义与价值。当前的云边协同资源调度机制已难以满足当前应用场景对任务处理的需求。
发明内容
本发明的目的是提供一种随机性在线云边端协同的数据存储方法及系统,以解决现有技术中的不足,它根据数据的历史存储行为以及云、边两侧对数据的处理频次,来实时提供一种智能、灵活、高效的数据存储策略,在保证数据存储性能的同时极大地降低云存储服务的部署成本,提高云存储服务的部署效率。
本申请的一个实施例提供了一种随机性在线云边端协同的数据存储方法,所述方法包括:
步骤1:开始选取研究对象,在一个地域范围内,随机选取一个可用区作为对象,统计该可用区AZ中边缘端节点的数量n。
步骤2:设定相关参数,令bcenter为云端存储的数据量,令为边缘端存储的数据量,令pcenter为数据在云端存储的单价,令为数据在边缘端存储的单价,其中i∈[1,n],令q为数据存取的单价,令t为数据的存储时长,令T为存储计费的周期。假设在过去一个存储周期中数据存放在云端,令其存储成本为Ccenter,假设在过去一个存储周期中数据存放在边缘端,令其存储成本为Cedge
步骤3:设定滑动窗口尺寸大小并统计相关数据,设定滑动窗口的尺寸大小z,实时监测终端设备的数据请求情况,记录并统计滑动窗口范围内云用户在终端发起的数据存取请求次数m,统计这些数据在云端和边缘端进行处理的次数,其中将在云端进行数据处理的次数记作hcenter,将在边缘端进行数据处理的次数记作hedge
步骤4:计算弹性平衡点βe,构造概率密度函数,获取随机因子k;
步骤5:设计随机性在线云边端协同数据存储算法;
步骤6:通过所述机性在线云边端协同数据存储算法判断当前数据是否需要进行迁移。
可选的,所述随机性在线云边端协同的数据存储方法还包括:
若所述当前数据需要迁移,则将存储在边缘端的数据迁移到云端,或将云端的数据迁移到边缘端,在当前存储周期T,随机性在线云边端协同的数据存储方法执行完毕,在下一存储周期T',重复执行该方法,若所述当前数据不需要迁移,则直接结束进程。
进一步的,所述计算弹性平衡点βe是为提出随机性在线云边端协同数据存储算法做前期准备,所述弹性平衡点βe通过步骤1-3中设定的相关参数,得出在过去一个存储计费周期T中,数据存储在云端的成本为Ccenter=bcenter·pcenter·t+z·hcenter/T·q,数据存储在边缘端的成本为
进一步的,所述计算弹性平衡点βe中将云端数据的存储成本Ccenter和边缘端数据的存储成本Cedge建立关联,并设定弹性区间s为[l,r],其中l=min(z/T·hcenter,m),r=max(z/T·hedge,m)。
进一步的,所述弹性平衡点βe通过计算得出:
进一步的,在步骤4中,所述构造概率密度函数通过结合数据存储成本的计算公式以及设定的相关参数再结合计算得出的弹性平衡点βe,其中令 所述构造概率密度函数如下:
其中,e为自然指数,δ(·)为狄拉克δ函数,k为随机因子。
进一步的,所述构造的概率密度函数,通过计算其反函数来获取随机因子k,其中令函数f(y)的数学期望计算f(k)的反函数得到随机因子k,其随机因子k的计算公式为:
进一步的,在步骤5中,所述设计随机性在线云边端协同数据存储算法,包括如下步骤:
在当前时刻t0,分别统计从t0-z+1时刻到t0时刻期间数据在云端和边缘端被访问的次数hcenter和hedge,并由此计算出这段时间数据的存储成本
如果当前的数据存储在边缘端,在当前时刻t0比较数据的存储成本和随机因子k的大小;
如果当前的数据存储在云端,在当前时刻t0比较数据的存储成本和随机因子k的大小;
设置多级队列协调机制,创建n级存储队列,在当前时刻t0,通过比较数据的存储成本和随机因子k的大小,确定数据应当存储在云端还是边缘端,对于需要进行物理迁移的数据,基于局部性原理首先将其存放到队列ni中进行缓存,然后完成数据的迁移。
本申请的又一实施例提供了一种随机性在线云边端协同的数据存储系统,所述系统包括:
数据采集模块,获取云端、边缘端和终端的资源请求信息,并接收与服务器连接的数据存取记录信息;
实时监控模块,实时监控云端、边缘端和终端的数据存取请求,包括请求时间、数据量大小等参数;
数据存储模块,对采集的数据、设定的参数以及实时监控的信息进行存储;
计算模块,计算弹性平衡点、构造概率密度函数、获取随机因子、计算反函数、计算数学期望等关键数据;
算法生成模块,生成随机性在线云边端协同数据存储算法,判断当前数据应当存储在云端还是边缘端;
协同调度模块,将存储在云端的数据迁移到边缘端,或将存储在边缘端的数据迁移到云端;
日志模块,记录终端用户资源请求情况以及数据获取情况等信息,并对云端和边缘端两侧数据的迁移情况进行记录。
本申请的又一实施例提供了一种存储介质,用于存储计算机指令,所述计算机指令被处理器执行时,完成随机性在线云边端协同的数据存储方法。
本申请的又一实施例提供了一种终端设备,包括存储器、处理器和计算机程序,所述存储器中存储有计算机程序,所述处理器被设置为运行所述计算机程序以实现上述任一项中所述的方法。
与现有技术相比,本发明可以实时地为数据应当存储在边缘端还是云端做出合理的决策,并尽可能降低云存储服务的部署成本,提高云存储服务的部署效率,同时能够保证用户在使用云存储服务时的QoS。提出了一种随机性在线云边端协同数据存储算法,通过将随机性在线云边端协同数据存储算法与概率分布、数学期望相结合,可以尽可能地降低云存储服务的部署成本,随机性在线云边端协同数据存储算法产生的随机因子可以实时地为数据应当存储在边缘端还是云端做出更加合理的决策。引入“滑动窗口”的概念,实现对存储在云端和边缘端的数据实时监控,并支持自定义滑动窗口的尺寸大小,可以自适 应地调整云端和边缘端对所存储数据的感知范围,通过将随机性在线云边端协同数据存储算法与滑动窗口相结合,可以有效提升随机性在线云边端协同数据存储算法预测结果的准确性与灵活性,并能够轻松应对不同的生产环境需求,提高云存储服务的部署效率,提出“弹性平衡点”(elastic-break-even-points)的概念,通常情况下,平衡点都是基于某一等式计算得出的定值,这样可以简化研究模型。然而,在一些对时延感知较为敏感的应用场景,设计一个具有弹性区间的平衡点,能够为数据应当存储在边缘端还是云端做出更加合理的决策,并在保证数据存储性能的同时极大地降低数据存储成本。引入多级队列协调机制,通过设定多级队列,基于局部性原理实现对云端和边缘端所存储数据的优先级缓存,从而在数据存取、使用的过程中更加稳定,以充分保证用户在使用云存储服务时的QoS。
附图说明
图1为本发明实施例一中提供的随机性在线云边端协同的数据存储方法流程示意图;
图2为本发明实施例一中提供的随机性在线云边端协同数据存储算法流程示意图;
图3为本发明实施例二中提供的一种随机性在线云边端协同的数据存储系统架构结构图。
具体实施方式
下面通过参考附图描述的实施例是示例性的,仅用于解释本发明,而不能解释为对本发明的限制。
本发明实施例首先提供了一种基于富文本编辑器的编辑信息纠错方法,该方法可以应用于终端设备,如计算机终端,具体如普通电脑、量子计算机等。
下面以运行在计算机终端上为例对其进行详细说明。
实施例1:
图1为本发明实施例提供的一种随机性在线云边端协同的数据存储方法,包括如下步骤:
步骤1:选取研究对象。在一个地域(Region)范围内,随机选取一个可用区(Availability Zone)作为对象,统计该可用区AZ中边缘端节点的数量n。
步骤2:设定相关参数。令bcenter为云端存储的数据量,令为边缘端存储的数据量,令pcenter为数据在云端存储的单价,令为数据在边缘端存储的单价,其中i∈[1,n],令q为数据存取的单价,令t为数据的存储时长,令T为存储计费的周期。假设在过去一个存储周期中数据存放在云端,令其存储成本为Ccenter,假设在过去一个存储周期中数据存放在边缘端,令其存储成本为Cedge
步骤3:设定滑动窗口尺寸大小并统计相关数据。设定滑动窗口的尺寸大小z,实时监测终端设备的数据请求情况。记录并统计滑动窗口范围内云用户在终端发起的数据存取请求次数m。统计这些数据在云端和边缘端进行处理的次数。其中,将在云端进行数据处理的次数记作hcenter,将在边缘端进行数据处理的次数记作hedge
步骤4:通过概率论、反函数等进行数学运算,计算弹性平衡点βe,构造概率密度函数,获取随机因子k,为接下来提出随机性在线云边端协同数据存储算法做前期准备。
弹性平衡点βe
通过步骤1-3中设定的相关参数,可以得出在过去一个存储计费周期T中,数据存储在云端的成本为Ccenter=bcenter·pcenter·t+z·hcenter/T·q,数据存储在边缘端的成本为
进一步地,将云端数据的存储成本Ccenter和边缘端数据的存储成本Cedge建立关联,并设定弹性区间s为[l,r]。其中,l=min(z/T·hcenter,m),r=max(z/T·hedge,m),进而计算出弹性平衡点βe
概率密度函数:
结合数据存储成本的计算公式以及设定的相关参数,结合计算得出的弹性平衡点βe,令构造如下的概率密度函数。
其中,e为自然指数,δ(·)为狄拉克δ函数,k为随机因子。
随机因子k:
通过构造的概率密度函数,计算其反函数,并通过反函数来获取随机因子k。令函数f(y)的数学期望计算f(k)的反函数,可以得到随机因子k。令F(k)=u,则随机因子k的计算公式为:
步骤5:随机性在线云边端协同数据存储算法,其执行流程如图2所示,包括如下步骤:
在当前时刻t0,分别统计从t0-z+1时刻到t0时刻期间数据在云端和边缘端被访问的次数hcenter和hedge,并由此计算出这段时间数据的存储成本
如果当前的数据存储在边缘端,在当前时刻t0比较数据的存储成本和随机因子k的大小,若那么数据存储在云端会节约更多的成本;若那么数据继续存储在边缘端会节约更多的成本。
如果当前的数据存储在云端,在当前时刻t0比较数据的存储成本和随机因子k的大小,若那么数据存储在边缘端会节约更多的成本;若那么数据继续存储在云端会节约更多的成本。
设置多级队列协调机制,创建n级存储队列。在当前时刻t0,通过比较数据的存储成本和随机因子k的大小,确定数据应当存储在云端还是边缘端。对于需要进行物理迁移的数据,基于局部性原理首先将其存放到队列ni中进行缓存,然后完成数据的迁移。
步骤6:在当前存储周期T,随机性在线云边端协同的数据存储方法执行完毕。在下一存储周期T',重复执行该方法。
本实施例详细描述了云端冷热数据随机性在线分层存储方法的一种实施方式,用户在使用本公开所述方法时,不应局限于本实施例所述的方式,可以根据自己的业务和实际情况进行适当调整。
实施例2:
参见图3,图3为本发明实施例提供的一种随机性在线云边端协同的数据存储系统,本系统包括七个模块,分别为:数据采集模块、实时监控模块、数据存储模块、计算模块、算法生成模块、协同调度模块和日志模块。
具体的,数据采集模块能够获取云端、边缘端和终端的资源请求信息,并接收与服务器连接的数据存取记录信息;实时监控模块能够实时监控云端、边缘端和终端的数据存取请求,包括请求时间、数据量大小等参数;数据存储模块负责对采集的数据、设定的参数以及实时监控的信息进行存储;计算模块主要用于计算弹性平衡点(elastic-break-even-point)、构造概率密度函数、获取随机因子、计算反函数、计算数学期望等关键数据,为生成随机性在线云边端协同数据存储算法提供支持;算法生成模块通过生成随机性在线云边端协 同数据存储算法,判断当前数据应当存储在云端还是边缘端;协同调度模块,负责将存储在云端的数据迁移到边缘端,或将存储在边缘端的数据迁移到云端;日志模块,用于记录终端用户资源请求情况以及数据获取情况等信息,并对云端和边缘端两侧数据的迁移情况进行记录。
实施例3:
本实施例还提供了一种电子设备,包括:一个或多个处理器、一个或多个存储器、以及一个或多个计算机程序;其中,处理器与存储器连接,上述一个或多个计算机程序被存储在存储器中,当电子设备运行时,该处理器执行该存储器存储的一个或多个计算机程序,以使电子设备执行上述实施例一所述的方法。
需要说明的是,本实施例中处理器可以是中央处理单元CPU,处理器还可以是其他通用处理器、数字信号处理器DSP、专用集成电路ASIC,现成可编程门阵列FPGA或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等,通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
更具体的,存储器可以包括只读存储器和随机存取存储器,并向处理器提供指令和数据、存储器的一部分还可以包括非易失性随机存储器。例如,存储器还可以存储设备类型的信息。
在实现过程中,实施例1的方法各步骤可以通过处理器中的硬件的集成逻辑电路或者软件形式的指令完成。
在更多实施例中,还提供:
一种计算机可读存储介质,用于存储计算机指令,所述计算机指令被处理器执行时,完成实施例一所述的方法。
本领域普通技术人员可以意识到,结合本实施例描述的各示例的单元及算法步骤,能够以电子硬件或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行,取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能,但 是这种实现不应认为超出本申请的范围。
以上对本发明实施例进行了详细介绍,本文中应用了具体个例对本发明的原理及实施方式进行了阐述,以上实施例的说明只是用于帮助理解本发明的方法及其核心思想;同时对于本领域的一般技术人员,依据本发明的思想,在具体实施方式及应用范围上均会有改变之处,综上所述,本说明书内容不应理解为对本发明的限制。

Claims (10)

  1. 一种随机性在线云边端协同的数据存储方法,其特征在于,所述方法包括:
    步骤1:开始选取研究对象,在一个地域范围内,随机选取一个可用区作为对象,统计该可用区AZ中边缘端节点的数量n。
    步骤2:设定相关参数,令bcenter为云端存储的数据量,令为边缘端存储的数据量,令pcenter为数据在云端存储的单价,令为数据在边缘端存储的单价,其中i∈[1,n],令q为数据存取的单价,令t为数据的存储时长,令T为存储计费的周期。假设在过去一个存储周期中数据存放在云端,令其存储成本为Ccenter,假设在过去一个存储周期中数据存放在边缘端,令其存储成本为Cedge
    步骤3:设定滑动窗口尺寸大小并统计相关数据,设定滑动窗口的尺寸大小z,实时监测终端设备的数据请求情况,记录并统计滑动窗口范围内云用户在终端发起的数据存取请求次数m,统计这些数据在云端和边缘端进行处理的次数,其中将在云端进行数据处理的次数记作hcenter,将在边缘端进行数据处理的次数记作hedge
    步骤4:计算弹性平衡点βe,构造概率密度函数,获取随机因子k;
    步骤5:设计随机性在线云边端协同数据存储算法;
    步骤6:通过所述机性在线云边端协同数据存储算法判断当前数据是否需要进行迁移。
  2. 根据权利要求1所述的一种随机性在线云边端协同的数据存储方法,其特征在于,所述方法还包括:
    若所述当前数据需要迁移,则将存储在边缘端的数据迁移到云端,或将云端的数据迁移到边缘端,在当前存储周期T,随机性在线云边端协同的数据存储方法执行完毕,在下一存储周期T',重复执行该方法,若所述当前数据不需要迁移,则直接结束进程。
  3. 根据权利要求1所述的一种随机性在线云边端协同的数据存储方法,其特征在于,所述计算弹性平衡点βe是为提出随机性在线云边端协同数据存储算法做前期准备,所述弹性平衡点βe通过步骤1-3中设定的相关参数,得出在过去一个存储计费周期T中,数据存储在云端的成本为Ccenter=bcenter·pcenter·t+z·hcenter/T·q,数据存储在边缘端的成本为
  4. 根据权利要求3所述的一种随机性在线云边端协同的数据存储方法,其特征在于,所述计算弹性平衡点βe中将云端数据的存储成本Ccenter和边缘端数据的存储成本Cedge建立关联,并设定弹性区间s为[l,r],其中l=min(z/T·hcenter,m),r=max(z/T·hedge,m)。
  5. 根据权利要求4一种随机性在线云边端协同的数据存储方法,其特征在于,所述弹性平衡点βe通过计算得出:
  6. 根据权利要求1所述的一种随机性在线云边端协同的数据存储方法,其特征在于,在步骤4中,所述构造概率密度函数通过结合数据存储成本的计算公式以及设定的相关参数再结合计算得出的弹性平衡点βe,其中令 所述构造概率密度函数如下:
    其中,e为自然指数,δ(·)为狄拉克δ函数,k为随机因子。
  7. 根据权利要求1所述的一种随机性在线云边端协同的数据存储方法,其特征在于,所述构造的概率密度函数,通过计算其反函数来获取随机因子k,其中令函数f(y)的数学期望计算f(k)的反函数得到随机因子k,其随机因子k的计算公式为:
  8. 根据权利要求1所述的一种随机性在线云边端协同的数据存储方法,其特征在于,在步骤5中,所述设计随机性在线云边端协同数据存储算法,包括如下步骤:
    在当前时刻t0,分别统计从t0-z+1时刻到t0时刻期间数据在云端和边缘端被访问的次数hcenter和hedge,并由此计算出这段时间数据的存储成本
    如果当前的数据存储在边缘端,在当前时刻t0比较数据的存储成本和随机因子k的大小;
    如果当前的数据存储在云端,在当前时刻t0比较数据的存储成本和随机因子k的大小;
    设置多级队列协调机制,创建n级存储队列,在当前时刻t0,通过比较数据的存储成本和随机因子k的大小,确定数据应当存储在云端还是边缘端,对于需要进行物理迁移的数据,基于局部性原理首先将其存放到队列ni中进行缓存,然后完成数据的迁移。
  9. 一种随机性在线云边端协同的数据存储系统,其特征在于,所述系统包括七个模块:
    数据采集模块,获取云端、边缘端和终端的资源请求信息,并接收与服务器连接的数据存取记录信息;
    实时监控模块,实时监控云端、边缘端和终端的数据存取请求,包括请求时间、数据量大小等参数;
    数据存储模块,对采集的数据、设定的参数以及实时监控的信息进行存储;
    计算模块,计算弹性平衡点、构造概率密度函数、获取随机因子、计算反函数、计算数学期望等关键数据;
    算法生成模块,生成随机性在线云边端协同数据存储算法,判断当前数据应当存储在云端还是边缘端;
    协同调度模块,将存储在云端的数据迁移到边缘端,或将存储在边缘端的数据迁移到云端;
    日志模块,记录终端用户资源请求情况以及数据获取情况等信息,并对云端和边缘端两侧数据的迁移情况进行记录。
  10. 一种终端设备,包括:至少一个处理器、至少一个存储器、以及至少一个计算机程序,其特征在于,所述处理器与存储器连接,所述存储器中存储有计算机程序,所述处理器被设置为执行所述计算机程序以实现所述权利要求1至8任一项中所述的方法。
PCT/CN2023/142749 2023-12-13 2023-12-28 一种随机性在线云边端协同的数据存储方法及系统 Pending WO2024222020A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202311712874.5A CN117873377B (zh) 2023-12-13 2023-12-13 一种随机性在线云边端协同的数据存储方法及系统
CN202311712874.5 2023-12-13

Publications (1)

Publication Number Publication Date
WO2024222020A1 true WO2024222020A1 (zh) 2024-10-31

Family

ID=90580017

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/142749 Pending WO2024222020A1 (zh) 2023-12-13 2023-12-28 一种随机性在线云边端协同的数据存储方法及系统

Country Status (2)

Country Link
CN (1) CN117873377B (zh)
WO (1) WO2024222020A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN119903552A (zh) * 2024-12-31 2025-04-29 华中科技大学 一种云边融合的安全存储系统
CN120335967A (zh) * 2025-04-17 2025-07-18 南方电网综合能源股份有限公司 基于边缘计算的智能终端数据处理方法及系统
CN120614354A (zh) * 2025-06-25 2025-09-09 江苏悦达网络科技有限公司 一种基于边缘计算的云边协同管理方法及系统
CN120909531A (zh) * 2025-10-10 2025-11-07 江苏翌旭网络科技有限公司 基于云资源的数据分层存储迁移服务系统
CN121078049A (zh) * 2025-11-06 2025-12-05 成都通广网联科技有限公司 一种应用于高速公路的智能驾驶边缘协同计算系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111800486A (zh) * 2020-06-22 2020-10-20 山东大学 一种云边协同的资源调度方法及系统
CN112926262A (zh) * 2021-02-18 2021-06-08 同济大学 云边协同环境下的数据分存方法、系统、介质及终端
CN113326002A (zh) * 2021-05-22 2021-08-31 清华大学 基于计算迁移的云边协同控制系统及迁移决策生成方法
WO2022095022A1 (zh) * 2020-11-04 2022-05-12 网络通信与安全紫金山实验室 计算资源的调度方法、装置、计算机设备及存储介质
CN114564312A (zh) * 2022-03-03 2022-05-31 北京航空航天大学云南创新研究院 一种基于自适应深度神经网络的云边端协同计算方法
WO2023030513A1 (zh) * 2021-09-05 2023-03-09 汉熵通信有限公司 物联网系统

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20220019367A1 (en) * 2017-04-21 2022-01-20 Pure Storage, Inc. Migrating Data In And Out Of Cloud Environments
US11740821B2 (en) * 2021-04-12 2023-08-29 EMC IP Holding Company LLC Cost-aware garbage collection for cloud storage
CN114327853A (zh) * 2021-10-09 2022-04-12 北京航空航天大学 云边混合系统中面向复杂任务的低成本用户关联和计算迁移方法
CN115883371B (zh) * 2022-11-26 2024-09-03 重庆理工大学 边缘-云协同系统中基于学习优化方法的虚拟网络功能放置方法
CN116896569A (zh) * 2023-06-26 2023-10-17 山东大学 基于公有云存储服务的数据云边协同管理方法及系统

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111800486A (zh) * 2020-06-22 2020-10-20 山东大学 一种云边协同的资源调度方法及系统
WO2022095022A1 (zh) * 2020-11-04 2022-05-12 网络通信与安全紫金山实验室 计算资源的调度方法、装置、计算机设备及存储介质
CN112926262A (zh) * 2021-02-18 2021-06-08 同济大学 云边协同环境下的数据分存方法、系统、介质及终端
CN113326002A (zh) * 2021-05-22 2021-08-31 清华大学 基于计算迁移的云边协同控制系统及迁移决策生成方法
WO2023030513A1 (zh) * 2021-09-05 2023-03-09 汉熵通信有限公司 物联网系统
CN114564312A (zh) * 2022-03-03 2022-05-31 北京航空航天大学云南创新研究院 一种基于自适应深度神经网络的云边端协同计算方法

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN119903552A (zh) * 2024-12-31 2025-04-29 华中科技大学 一种云边融合的安全存储系统
CN120335967A (zh) * 2025-04-17 2025-07-18 南方电网综合能源股份有限公司 基于边缘计算的智能终端数据处理方法及系统
CN120614354A (zh) * 2025-06-25 2025-09-09 江苏悦达网络科技有限公司 一种基于边缘计算的云边协同管理方法及系统
CN120909531A (zh) * 2025-10-10 2025-11-07 江苏翌旭网络科技有限公司 基于云资源的数据分层存储迁移服务系统
CN121078049A (zh) * 2025-11-06 2025-12-05 成都通广网联科技有限公司 一种应用于高速公路的智能驾驶边缘协同计算系统

Also Published As

Publication number Publication date
CN117873377A (zh) 2024-04-12
CN117873377B (zh) 2025-11-11

Similar Documents

Publication Publication Date Title
CN108509276B (zh) 一种边缘计算环境中的视频任务动态迁移方法
CN117873377B (zh) 一种随机性在线云边端协同的数据存储方法及系统
CN112000703B (zh) 数据入库处理方法、装置、计算机设备和存储介质
CN103999049B (zh) 用于预测虚拟机需求的方法和装置
CN103617087B (zh) 一种适合迭代计算的MapReduce优化方法
Li An adaptive overload threshold selection process using Markov decision processes of virtual machine in cloud data center
CN112486653B (zh) 调度多类型计算资源的方法、装置和系统
WO2021159638A1 (zh) 集群队列资源的调度方法、装置、设备及存储介质
CN117972367B (zh) 一种数据存储预测方法、数据存储子系统及智能计算平台
US20130268941A1 (en) Determining an allocation of resources to assign to jobs of a program
CN106101196B (zh) 一种基于概率模型的云渲染平台任务调度系统
Li et al. Flink‐ER: An Elastic Resource‐Scheduling Strategy for Processing Fluctuating Mobile Stream Data on Flink
CN106126340A (zh) 一种跨数据中心云计算系统的规约器选择方法
Qin et al. Joint energy optimization on the server and network sides for geo-distributed data centers
Fan et al. Graph4Edge: a graph-based computation offloading strategy for mobile-edge workflow applications
CN119149244A (zh) 算力调度方法及装置
Huang et al. Computation offloading for multimedia workflows with deadline constraints in cloudlet-based mobile cloud
CN110308991B (zh) 一种基于随机任务的数据中心节能优化方法及系统
Alkhalaileh et al. Dynamic resource allocation in hybrid mobile cloud computing for data-intensive applications
Li et al. Toward optimal operator parallelism for stream processing topology with limited buffers
CN106209975A (zh) 一种跨数据中心云计算系统的资源供给方法
Casalicchio et al. An energy-aware adaptation model for big data platforms
Trabelsi et al. Application topology definition and tasks mapping for efficient use of heterogeneous resources
CN111736968B (zh) 配电系统云数据中心计算资源优化配置方法、装置、设备及介质
Tang et al. To cloud or not to cloud: an on-line scheduler for dynamic privacy-protection of deep learning workload on edge devices

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23935186

Country of ref document: EP

Kind code of ref document: A1