International Business Machines Corporation

Nathaniel S Tomsic

Vasu Vallabhaneni

Khalid Filali-Adib

Neal Richard Marion

David W Sheffield

Maria Garza

Sungjin Yook

Kevin Lynn Fought

Perinkulam I Ganesh

Stephen Bailey Peckham

Frederic Barrat

Marc Joel Stephenson

Laurent Dufour

Christine M Briand

J Mark McConaughy

According to one aspect of the present disclosure, a method and technique for mobility operation resource allocation is disclosed. The method includes: receiving a request to migrate a running application from a first machine to a second machine; displaying an adjustable resource allocation mobility setting interface indicating a plurality of mobility settings comprising at least one performance-based mobility setting and at least one concurrency-based mobility setting; receiving, via the interface, a selection of a mobility setting defining a resource allocation to utilize for the migration; and migrating the running application from the first machine to the second machine utilizing resources as set by the selected mobility setting.

Mobility operation resource allocation

A method, apparatus, and computer program product are provided in a data processing system for performing a logical partition migration utilizing multiple mover service partition pairs. Responsive to a virtual machine monitor initiating a logical partition migration operation to move a logical partition from a source system to a destination system, a plurality of input/output paths are established between a plurality of mover service partition pairs. The virtual machine monitor performs the logical partition migration operation using the plurality of mover service partition pairs to transfer a memory image of the logical partition from the source system to the destination system to effect the logical partition migration operation. Responsive to failure of one of the plurality of input/output paths, the virtual machine monitor may complete the logical partition migration operation using at least one remaining mover service partition pair.

Performing a logical partition migration utilizing plural mover service partition pairs

Provided are techniques for comparing a first fileset associated with a first operating system (OS) with a second fileset associated with a second OS; determining, based upon the comparing, that the second OS is a more current version of the first OS; in response to the determining that the second OS is a more current version of the first OS, moving, in conjunction with live application mobility, a virtual machine (VM) workload partition (WPAR) on the first LPAR to a second LPAR, the moving comprising determining a set of overlays associated with the WPAR corresponding to the second OS; removing from the WPAR any overlays associated with the first OS; applying to the WPAR a set of overlays corresponding to the second OS; check pointing processes associated with the WAPR; and copying live data associated with the LPAR from the first LPAR to the second LPAR.

Live application mobility from one operating system level to an updated operating system level and applying overlay files to the updated operating system

Provided are techniques for comparing a first fileset associated with a first operating system (OS) with a second fileset associated with a second OS; determining, based upon the comparing, that the second OS is a more current version of the first OS; in response to the determining that the second OS is a more current version of the first OS, moving, in conjunction with live application mobility, a virtual machine (VM) workload partition (WPAR) on the first LPAR to a second LPAR, the moving comprising determining a set of overlays associated with the WPAR corresponding to the second OS; removing from the WPAR any overlays associated with the first OS; applying to the WPAR a set of overlays corresponding to the second OS; check pointing processes associated with the WPAR; and copying live data associated with the LPAR from the first LPAR to the second LPAR.

A mechanism is provided in a data processing system for performing a logical partition migration utilizing multiple paths. Responsive to a virtual machine monitor initiating a logical partition migration operation to move a logical partition from a source system to a destination system, the mechanism allocates a plurality of memory pools and a plurality of threads from a mover service partition to the virtual machine monitor. The virtual machine monitor performs the logical partition migration operation utilizing each of the plurality of threads to transfer a corresponding one of the plurality of memory pools from the source system to the destination system to effect the logical partition migration operation.

Utilizing multiple memory pools during mobility operations

According to one aspect of the present disclosure a system and technique for mobility operation resource allocation is disclosed. The system includes a first machine having a running application, and a management console operable to manage migrating the running application from the first machine to a second machine. The management console is operable to: display an interface indicating a plurality of selectable mobility settings, each mobility setting corresponding to a desired resource allocation to utilize for the migration; receive via the interface, a selection of a mobility setting to apply for migrating the running application; and initiate migration of the running application from the first machine to the second machine utilizing resources as set by the selected mobility setting.

Some embodiments of the inventive subject matter are directed to receiving a request from a first instance of an operating system (e.g., a virtual operating system) to load a kernel extension that extends functionality of a kernel, where the kernel and the first instance of the operating system are managed by a second instance of the operating system (e.g., a global operating system), and where the first and second instances of the operating system share the kernel. Some embodiments are further directed to recording an indicator that indicates that the first of the plurality of the instances of the operating system requested to load the kernel extension, where the indicator is accessible only to the second instance of the operating system. In some embodiments, the method is further directed to loading the kernel extension, where loading the kernel extension extends functionality of the kernel. In some embodiments, the method is further directed to receiving a request by the first instance of the operating system to remove the functionality from the kernel and determining, based on the indicator, that the first instance of the operating system had previously requested to load the kernel extension. In some embodiments, the method is further directed to removing the functionality from the kernel in response to the determining, based on the indicator, that the first instance of the operating system had previously requested to load the kernel extension.

Tracking loading and unloading of kernel extensions in isolated virtual space

Some embodiments of the inventive subject matter are directed to detecting a file that a first instance of an operating system is configured to load where the file is configured to extend functionality of a kernel when loaded. In some embodiments, the first instance of the operating system is managed by a second instance of the operating system (e.g., the first instance of the operating system is a virtual operating system and the second instance is a global operating system). The first and second instances of the operating system share the kernel. Some embodiments are further directed to generating a unique electronic identifier based on properties of the file. The unique electronic identifier uniquely identifies the file. Some embodiments are further directed to providing a copy of the file to the first instance of the operating system, and some time later, receiving a request from the first instances of the operating system to load the copy of the file. Some embodiments are further directed to determining, using the unique electronic identifier, that the copy of the file from the first instance of the operating system is equivalent to the file. Some embodiments are further directed to loading the copy of the file in response to determining, using the unique electronic identifier, that the copy of the file is equivalent to the file. The loading extends the functionality of the kernel.

Managing unique electronic identification for kernel extensions in isolated virtual space

Some embodiments of the inventive subject matter are directed to receiving a request from a first instance of an operating system (e.g., a virtual operating system) to load a kernel extension that extends functionality of a kernel, where the kernel and the first instance of the operating system are managed by a second instance of the operating system (e.g., a global operating system), and where the first and second instances of the operating system share the kernel. Some embodiments are further directed to loading the kernel extension as a shared kernel extension. Some embodiments are further directed to receiving a request by the first instances of the operating system to unload the kernel extension and determining that a third instance of the operating system (e.g., another virtual operating system also managed by the global operating system) requires the kernel extension to remain loaded. Some embodiments are further directed to notifying the first instance of the operating system that the kernel extension was unloaded while leaving the kernel extension loaded in the kernel.

Growing community of inventors

Austin, TX, United States of America

Nathaniel S Tomsic