Hypervisor subpartition as concurrent upgrade

摘要

A processor-implemented method for a concurrent software service upgrade is provided. The processor implemented method may include receiving a type of service request corresponding to the software service upgrade, determining, by the processor, the type of service request and then generating a plurality of subpartitions corresponding to a hypervisor. The method may further include applying the service request to at least one subpartition within the plurality of subpartitions, wherein the service request is applied to the at least one subpartition based on the type of service request and balancing the system resources among the plurality of subpartitions upon the applying of the service request to the at least one subpartition.

主权项

1. A processor-implemented method for providing a concurrent software service upgrade, the method comprising:
receiving, by a management tool, a type of service request corresponding to the software service upgrade; determining, by the management tool, the type of service request; invoking, by the management tool, a control program subpartition configured to apply the service request to a plurality of subpartitions based on the type of service request, wherein the plurality of subpartitions includes at least one upgradeable micro hypervisor kernel module, micro hypervisor I/O offload subpartition and micro hypervisor host subpartition; applying the service request to at least one subpartition within the plurality of subpartitions, wherein applying the service request to at least one subpartition comprises:
applying the service request to a plurality of micro hypervisor kernels based on the determined type of service request being associated with at least one upgradeable kernel module;applying the service request to a micro hypervisor I/O offload subpartition based on the determined type of service request being associated with a service to an I/O offload subpartition, wherein applying the service request to the micro hypervisor I/O offload subpartition comprises generating a new I/O offload subpartition, applying the service to the generated new I/O offload subpartition, instructing an old I/O offload subpartition to stop taking new requests, determining when all existing requests are satisfied, and deleting the old offload subpartition, in response to determining that the existing requests are satisfied; andapplying the service request to a micro hypervisor host subpartition based on the determined type of service request not being associated with a service to an I/O offload subpartition and the determined type of service request not being associated with the at least one upgradeable kernel module, wherein applying the service request to the micro hypervisor host subpartition comprises generating a new micro hypervisor host subpartition, moving a plurality of virtual servers from an old host subpartition to the generated new micro hypervisor host subpartition, stopping the old host subpartition, and deleting the old host subpartition; and balancing a plurality of system resources among the plurality of subpartitions upon the applying of the service request to the at least one subpartition.