Scheduling method and scheduling end for FogCDN scene
A scheduling method and scene technology, applied in the field of shared computing, can solve problems such as discounts for the entire node and service capacity, and achieve the effects of stability and expansion, low cost, and bandwidth cost savings
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0151] Please refer to figure 1 , Embodiment 1 of the present invention is:
[0152] The optimal scheduling method in this embodiment is used in a FogCDN scenario, where the client refers to the client that requests specific content / data / resources / resource fragments; the server refers to the client that stores the corresponding content / data / resource / resource Therefore, a terminal node may be a client node, a server node, or both a client node and a server node. That is, this embodiment provides a requester node in a FogCDN scenario. and the scheduling optimization method between the server and the server. At the same time, for the existing business scenarios, in order to facilitate understanding, this embodiment is the probabilistic scheduling on the client. If the scheduling decision occurs on the server, the following C and S is exchanged with each other, and the corresponding meaning of matrix P also changes, and its element p ij Indicates the probability or proportion of...
Embodiment 2
[0162] Please refer to figure 1 , the second embodiment of the present invention is:
[0163] For the scheduling method used in the FogCDN scenario, on the basis of the first embodiment above, in this embodiment, the highest overall utilization of FogCDN nodes is the optimization goal, and the specific steps for obtaining P are:
[0164] Solve the first model to obtain the first scheduled probability moment P 1 , where the first model is:
[0165]
[0166] In this embodiment, matrix C is the distribution ratio vector of the number of client requests for a certain famous business served by PearShare under 5 node types, and matrix S is the statistics of all Fog nodes serving this business from PearShare in 5 types The distribution ratio vector under the node type, the matrix L is the connectivity rate matrix obtained through a lot of testing and practice under the condition of using some of the proprietary technologies of Lixiang, as follows:
[0167] S=[0.14 0 0.47 0.09 0...
Embodiment 3
[0175] Please refer to figure 1 , Embodiment three of the present invention is:
[0176] For the scheduling method used in the FogCDN scenario, on the basis of the second embodiment above, in this embodiment, the minimum overall connection delay of Fog nodes is the optimization goal, then P 1 After that also include steps:
[0177] Solve the second model to obtain the second scheduled probability matrix P 2 , the second model is:
[0178]
[0179] In this embodiment, the values of matrix S, matrix C, and matrix L are the same as those in Embodiment 2, and D is the connection delay matrix in the case of using a stronger "two-way hole punching" technology, and its value is:
[0180]
[0181] Simultaneously, The value range of is [1.5X,6X], X is when P=P 1 Time Target||S-C(P·L)|| 1 The value of , that is, according to Example 2, the optimal solution ||S-C(P·L)|| 1 = 0.0196948. In this example Take 0.03, that is, 97% of the Fog service node capabilities particip...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


