International Business Machines Corporation

Jacob Jason Rosales

James Anthony Pafumi

Morgan Jeffrey Rosas

Vasu Vallabhaneni

David Nevarez

Veena Ganti

Veena Patwari

Neal Richard Marion

Greg Robert Mewhinney

Basu Vaidyanathan

James P Allen

George James Romano

Robert G Kovacs

Michael Paul Cyr

Stephen M Tee

Maria Garza

James Britt Partridge

Ray W Anderson

Carl Duane Bender

A method utilizes cluster-awareness to effectively support a live partition mobility (LPM) event and provide recovery from node failure within a Virtual Input/Output (I/O) Server (VIOS) cluster. An LPM utility creates a monitoring thread on a first VIOS on initiation of a corresponding LPM event. The monitoring thread tracks a status of an LPM and records status information in the mobility table of a database. The LPM utility creates other monitoring threads on other VIOSes running on the (same) source server. If the first VIOS VIOS sustains one of multiple failures, the LPM utility provides notification to other functioning nodes/VIOSes. The LPM utility enables a functioning monitoring thread to update the LPM status. In particular, a last monitoring thread may perform cleanup/update operations within the database based on an indication that there are nodes on the first server that are in failed state.

Multiple node/virtual input/output (I/O) server (VIOS) failure recovery in clustered partition mobility

A method, data processing system and computer program product provide scalable data synchronization for a virtual input/output server (VIOS) cluster and one or more registered callers. A first VIOS is commits as a primary node of the VIOS cluster and performs the functions of: registering one or more callers to receive notification from the first VIOS of specific events occurring within the cluster; receiving notification of an occurrence of one of the specific events; and in response to receiving notification of the specific events, a deamon of the first VIOS retrieving a message payload file from a message payload file directory within the shared VIOS DB and passing the message payload file to the API, which forwards/posts the relevant event notification information from the message payload file to the TCP socket of each registered caller.

Single point, scalable data synchronization for management of a virtual input/output server cluster

A method, system, and computer program product provides effective management and deployment of image files within a Virtual Input/Output (I/O) Server (VIOS) cluster. An Image Management (IM) utility utilizes a discovery process to establish connections with objects in order to manage these objects. Following discovery of an ICP client, the IM utility enables creation of a virtual server partition and initiates the importing of a captured image to a VIOS cluster Storage Pool (SP). The IM utility creates logical units (LUs) within the Image Management Fileset (IMF) for each file within the captured image. The IM utility maps the LUs to the VIOS Cluster SP using unique image file handles. The IM utility configures the new storage that comprises the created LUs. The IM utility copies image files to the appropriate storage device/LU. When import/copy of the image is completed, the IM utility activates the virtual server partition.

Management of file images in a virtual environment

Various systems, processes, and products may be used to update virtual machines. In particular implementations, a system, process, and product for updating virtual machines may include the ability to determine whether a change to a portion of an operating system for a virtual machine is available and identify a virtual machine using the operating system. The system, process, and product may also include the ability to determine when the virtual machine is modifiable and update the portion of the operating system for the virtual machine when it is modifiable.

Virtual machine updates

In a data processing system having a plurality of virtualized operating system (OS) partitions, a first cluster-aware virtual input/output server (VIOS) is provided within an OS partition by execution of a cluster-aware (CA) OS on a virtual processor resource of the first VIOS. The CA OS establishes the VIOS as a cluster-aware VIOS by registering the first VIOS with a VIOS cluster comprising a second VIOS. Registering the first VIOS to the VIOS cluster enables the first VIOS to receive VIOS cluster configuration data and status data, which provides the first VIOS with information about the other VIOSes within the VIOS cluster. The first VIOS is thus able to communicate with other VIOSes within the VIOS cluster and share I/O resources with the other VIOSes within the VIOS cluster. Relevant cluster information is stored within a local storage of the first VIOS.

Registration and initialization of cluster-aware virtual input/output server nodes

In a data processing system including a virtual I/O server (VIOS) cluster and multiple logical partitions (LPARs), at least one VIOS of the VIOS cluster performs functions of: receiving first registration information from one or more entities within the VIOS cluster; registering, based on the first registration information, a handler associated with a first message type; responsive to receiving second registration information, registering, based on the second registration information, a listener associated with a second message type; receiving a first message associated with a first alert event of the cluster; determining, based on the first message, that the handler is associated with the first message type; and calling the handler. In one embodiment, the handler is associated with a pointer to a subroutine, and registering the handler includes storing the pointer in a data structure that associates the handler with the first message type.

VIOS cluster alert framework

A first virtual I/O server (VIOS) provides a cluster aware (CA) operating system (OS) executing on a processor resource of the first VIOS to register the first VIOS within a VIOS cluster. The first VIOS comprises a first field/failure data capture (FFDC) module that executes within the first VIOS and performs the functions of: receiving from an event listener a signal indicating that an FFFDC event/condition has been detected by the first VIOS; and automatically transmitting FFDC data to the shared storage repository for storage of the FFDC data within the shared storage repository. The FFDC module further performs the functions of: transmitting to one or more second VIOSes within the VIOS cluster, one or more messages to inform the one or more second VIOSes of an occurrence of the FFDC event/condition that was detected by the first VIOS.

Providing first field data capture in a virtual input/output server (VIOS) cluster environment with cluster-aware vioses

In a data processing system having a plurality of virtual input/output servers (VIOSes) configured within a VIOS cluster, a method, data processing system and computer program product provide for autonomous election of a primary node within a virtual input/output server (VIOS) cluster. A first VIOS performs the functions of: detecting that a primary node is required for the VIOS cluster; and autonomously initiating an election process to elect a next primary node from among the VIOSes within the VIOS cluster. When the first VIOS meets the pre-established requirements for becoming a primary node, the first VIOS obtains a lock on a primary node ID field within a VIOS database (DB) and then initiates a primary node commit process to assign the first VIOS as the primary node. The first VIOS issues a notification to the VIOS cluster to notify the other VIOSes that a primary node has been elected.

Autonomous primary node election within a virtual input/output server cluster

A method, data processing system, and computer program product autonomously migrate clients serviced by a first VIOS to other VIOSes in the event of a VIOS cluster 'split-brain' scenario generating a primary sub-cluster and a secondary sub-cluster, where the first VIOS is in the secondary sub-cluster. The VIOSes in the cluster continually exchange keep-alive information to provide each VIOS with an up-to-date status of other VIOSes within the cluster and to notify the VIOSes when one or more nodes loose connection to or are no longer communicating with other nodes within the cluster, as occurs with a cluster split-brain event/condition. When this event is detected, a first sub-cluster assumes a primary sub-cluster role and one or more clients served by one or more VIOSes within the secondary sub-cluster are autonomously migrated to other VIOSes in the primary sub-cluster, thus minimizing downtime for clients previously served by the unavailable/uncommunicative VIOSes.

Supporting autonomous live partition mobility during a cluster split-brained condition

A system and computer program product utilizes cluster-awareness to effectively support a live partition mobility (LPM) event and provide recovery from node failure within a Virtual Input/Output (I/O) Server (VIOS) cluster. An LPM utility creates a monitoring thread on a first VIOS on initiation of a corresponding LPM event. The monitoring thread tracks a status of an LPM and records status information in the mobility table of a database. The LPM utility creates other monitoring threads on other VIOSes running on the (same) source server. If the first VIOS sustains one of multiple failures, the LPM utility provides notification to other functioning nodes/VIOSes. The LPM utility enables a functioning monitoring thread to update the LPM status. In particular, a last monitoring thread may perform cleanup/update operations within the database based on an indication that there are nodes on the first server that are in failed state.

A method provides input/output (I/O) redundancy within a data processing system having (a) a client logical partition (LPAR) that generates and consumes I/O requests, (b) a plurality of virtual input/output servers (VIOS) that are communicatively inter-connected with each other to form a VIOS cluster and which include virtual I/O adapters for connecting to a fabric that provides access to a block storage. In one embodiment, a first VIOS receives an I/O request from the client LPAR. The first VIOS detects that a problem exists with a fabric connection to the block storage, and the first VIOS responds to the detected connection problem by autonomously propagating the I/O request to a second VIOS to which the first VIOS is connected. Forwarding of the I/O request to the block storage is subsequently completed by the second VIOS.

Autonomous propagation of virtual input/output (VIO) operation(s) to second VIO server (VIOS) due to a detected error condition at a first VIOS

In a data processing system including multiple logical partitions (LPARs), an application executes on a first logical partition (LPAR) of the multiple LPARs, where the application uses a first operation system stored in a first memory partition of a shared pool memory of the data processing system. A virtualization management component (a) initiates an update process that quiesces operations of the first LPAR, (b) pages in, via a virtual input/output server coupled to a first paging device, a first image of a second operating system from the first paging device to the shared pool memory; (c) changes one or more pointers associated with the application to point to one or more portions of the second operating system, such that the application uses the second operating system, when resumed; and (b) resumes execution the application.

Operating system image management

A method, system, and computer program product provide a shared virtual memory space via a cluster-aware virtual input/output (I/O) server (VIOS). The VIOS receives a paging file request from a first LPAR and thin-provisions a logical unit (LU) within the virtual memory space as a shared paging file of the same storage amount as the minimum required capacity. The VIOS also autonomously maintains a logical redundancy LU (redundant LU) as a real-time copy of the provisioned/allocated LU, where the redundant LU is a dynamic copy of the allocated LU that is autonomously updated responsive to any changes within the allocated LU. Responsive to a second VIOS attempting to read a LU currently utilized by a first VIOS, the read request is autonomously redirected to the logical redundancy LU. The redundant LU can be utilized to facilitate migration of a client LPAR to a different computing electronic complex (CEC).

Supporting virtual input/output (I/O) server (VIOS) active memory sharing in a cluster environment

A mechanism is provided for sharing an application among multiple server systems. In one embodiment, the application is stored on a fast disk paging device. The paging device is in communication with each of a plurality of servers so that each server may be given access to the application. Virtual real memory (VRM) is allocated to a first virtual client of the first server. A portion of this VRM is allocated to the application. The application is executed by a virtual client of a first server. During execution a condition may occur that requires taking the first server off line. In that event, VRM is allocated to a second virtual client of a second server. A portion of this VRM is allocated to the application. The application is then executed by the second virtual client of the second server.

Application mobility using virtual real memory services

Mechanisms are provided for detecting changes in virtual storage device configurations. The mechanisms detect an event corresponding to a change in configuration of a virtual storage device. The virtual storage device is comprised of a plurality of portions of a plurality of physical storage devices. The mechanisms further, in response to detecting the event, determine if the change in configuration of the virtual storage device results in a change in the types of physical storage devices that are part of the virtual storage device. Moreover, the mechanisms further transmit a notification, in response to a determination that the change in configuration of the virtual storage device results in a change in the types of physical storage devices that are part of the virtual storage device, of the results of the change in configuration of the virtual storage device to one or more registered recipients registered to receive such notifications.

Growing community of inventors

Austin, TX, United States of America

Jacob Jason Rosales