Cooperative scheduling method and system for computing resource and network resource of container cloud platform

A cooperative scheduling system for a computing resource and a network resource of a container cloud platform, comprising:

• a module for obtaining a load value of a container in a physical machine of a data center, configured to obtain a load value of a container in a physical machine of a data center, wherein each data center manages multiple physical machines, and each physical machine manages multiple containers;
• a current-container load margin calculation module, configured to calculate a load margin of a current container, wherein the load margin of the container is a difference between a rated load value of the container and a current load value of the container;
• a first determining module, configured to determine whether the load margin of the current container is less than 0;
• a first-container-sequence generation module, configured to:
• if the load margin of the current container is less than 0, calculate and compare load margins of other containers managed by a physical machine in which the current container is located, and sort, in descending order, the load margins of the other containers managed by the physical machine in which the current container is located, to generate a first container sequence;
• a second determining module, configured to determine whether a maximum load margin in the first container sequence is less than the load margin of the current container;
• a first migration module, configured to:
• if the maximum load margin is greater than or equal to the load margin of the current container, migrate load of the current container to a container corresponding to a load margin that is greater than 0 and that is in the first container sequence;
• wherein the first migration module specifically comprises:
• a first-migration-cost calculation unit, configured to calculate first migration costs for migrating the current container to all containers corresponding to load margins that are greater than 0 and that are in the first container sequence;
• a first comparison unit, configured to compare the first migration costs;
• and
• a first migration unit, configured to migrate the current container to a container corresponding to a lowest first migration cost;
• a second migration module, configured to:
• if the maximum load margin is less than the load margin of the current container, add a physical machine to a data management center in which the current container is located, and migrate load of the current container to a container managed by the added physical machine, or migrate load of the current container to a container managed by another physical machine in a data center in which the current container is located;
• wherein the second migration module specifically comprises:
• a total load margin calculation unit, configured to calculate a sum of load margins of all containers managed by each physical machine to obtain a total load margin of each physical machine;
• a second comparison unit, configured to compare and select a physical machine corresponding to a maximum total load margin;
• and
• a second migration unit, configured to migrate the load of the current container to a container managed by the physical machine corresponding to the maximum total load margin;
• a load margin calculation subunit, configured to calculate a load margin of each container managed by the physical machine corresponding to the maximum total load margin;
• a second-container-sequence generation subunit, configured to sort, in descending order, the load margins of the containers managed by the physical machine corresponding to the maximum total load margin, to generate a second container sequence;
• a second-migration-cost calculation subunit, configured to calculate second migration costs for migrating the current container to all containers corresponding to load margins that are greater than 0 and that are in the second container sequence;
• a second comparison subunit, configured to compare the second migration costs;
• and
• a second migration subunit, configured to migrate the current container to a container corresponding to a lowest second migration cost;
• a third determining module, configured to:
• when the load margin of the current container is greater than 0, determine whether to traverse all containers managed by a current physical machine;
• an end module, configured to:
• when all the containers managed by the current physical machine are traversed, end a scheduling task;
• and
• a container load margin update module, configured to:
• if not all the containers managed by the current physical machine have been traversed, obtain a load value of a next container managed by the current physical machine, calculate a load margin of the next container, update the calculated load margin of the next container to the load margin of the current container, and send an instruction to the first determining module.