Distributed node intra-group task scheduling method and system

A distributed node intra-group task scheduling system, the system comprising:

• a processor, configured to execute the following program modules stored in a memory:
• an information collecting module in each node in a group, configured to collect resource state information and history service information of the present node;
• a service command receiving module in a scheduling execution node in the group, configured to receive a service request command, and to parse the request command to obtain a request task;
• and
• a scheduling processing module in the scheduling execution node in the group, configured to:
• calculate a resource utilization rate increment after each node in the group loads a task and an occupation condition of each kind of resource in the group according to the resource state information and the history service information about each node;
• and
• select service execution nodes according to a principle of making the resource utilization rate increment as low as possible and the occupation of each kind of resource as balanced as possible, and provide the request task by the service execution nodes;
• the scheduling processing module further comprises:
• a first processing sub-module configured to calculate a resource occupation rate of a node according to the resource state information of the node by an equations of:
• rcj={c1,j,c2,j, . . . ,cN,j}, and
• ci,j=(si,j−soi,j)/si,j,
• wherein rcj represents a current resource occupation rate vector of any one node pj in the group, and ci,j represents an occupation rate of the i-th kind of resource for the node pj;
• a second processing sub-module configured to calculate an occupation rate of the resources over which the task O allocated to the node pj is executed;
• a third processing sub-module configured to calculate an occupation rate increment of the i-th kind of resource after the task O is loaded onto the node pj as compared with before it is loaded, and to further obtain a resource occupation rate increment vector of the task O for any one node pj in the group;
• a resource occupation influence rate obtaining sub-module configured to obtain a resource occupation influence rate of the node pj according to each element in the resource occupation rate increment vector;
• an intra-node resource balance degree obtaining sub-module configured to obtain an intra-node resource balance degree σj according to a variance among the occupation rates of different types of resources by an equation of:
• σ
• j
• =
• ∑
• i
• =
• 1
• N
• ⁢
• (
• ro
• i
• ,
• j
• -
• rt
• j
• _
• )
• 2
• N
• wherein
• rt
• j
• _
• ⁢
• ∑
• i
• =
• 1
• N
• ⁢
• ro
• i
• ,
• j
• N
• ,
• roi,j represents the occupation rate of the i-th kind of resource over which the node pj executes the task O, N represents the total number of types of resources supportable by all the nodes in the group of nodes, and i represents any one type of resource;
• an execution node set obtaining sub-module configured to determine nodes with lower resource balance degrees σj and lower resource occupation influence rates μj in the group of nodes as execution nodes to obtain an execution node set PK by an equation of:
• PK={pk}=argmin(μj,σj)T,∈[1,M′],
• wherein M′ represents the number of nodes in the eligible execution node set, and argmin represents a corresponding node which makes μj and σj get minimum values;
• and
• a selecting sub-module configured to select a node in the execution node set PK as a final execution node, and then forward the task O to the final execution node for processing.