CLUSTER-AWARE VIRTUAL INPUT/OUTPUT SERVER

• US 20120066678A1
• Filed: 09/10/2010
• Published: 03/15/2012
• Est. Priority Date: 09/10/2010
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Pafumi, James A., Rosales, Jacob J., Rosas, Morgan J., Vallabhaneni, Vasu
• Granted Patent: US 8,726,274 B2
• Time in Patent Office: Days
• US Class Current: 718/1
• CPC Class Codes: G06F 2009/45579 I/O management, e.g. providing access to device drivers or storage
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects

Abstract

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.

Claims

1. A computing electronic complex (CEC) comprising:

a processor;

one or more physical input/output (I/O) adapters that support I/O communication with an external network; and

a memory coupled to the one or more processors, wherein the memory includes;

a virtualization management component that supports creation of and communication between one or more virtualized operating system (OS) partitions;

at least one operating system (OS) partition, including a first virtual input/output (I/O) server (VIOS) having a cluster-aware (CA) OS that executes on a virtual processor resource of the VIOS to perform the function of;

registering the first VIOS with a VIOS cluster comprising at least a second VIOS, wherein the registering of the first VIOS to the VIOS cluster enables the first VIOS to receive cluster specific data to make the first VIOS aware of the VIOS cluster, wherein the first VIOS is able to communicate information with other VIOSes within the VIOS cluster.

2. The computing electronic complex of claim 1, wherein the registering of the first VIOS with the VIOS cluster comprises:

initializing a cluster registration module that assigns the VIOS with a unique ID and forwards VIOS registration data to a shared VIOS database (DB);

receiving, from the shared VIOS DB, cluster configuration data and status data of the VIOS cluster for local storage at the first VIOS; and

updating a local storage of the first VIOS with the received cluster configuration data and status data.

3. The computing electronic complex of claim 1, wherein the CA_OS further performs the functions of:

initializing an I/O emulation module of the CA_OS to enable the VIOS to provide virtual I/O (VIO) services to one or more client logical partitions (LPARs) existing within one of the OS partitions;

registering a first client LPAR with the VIOS;

assigning a unique VIO adapter to the client LPAR; and

enabling I/O operations for the client LPAR by linking the unique VIO adapter to one of the one or more physical I/O adapters.

4. The computing electronic complex of claim 1, wherein the CA_OS comprises a cluster management (CM) utility that executes on the virtual processor resource to provide the functions of:

registering the first VIOS with the VIOS DB for receipt of one or more notifications of one or more events that may occur n the VIOS cluster;

receiving events data from within the VIOS cluster; and

updating the local storage of the first VIOS with relevant cluster information for the first VIOS.

5. The computing electronic complex of claim 4, wherein the CM utility further provides the function of coordinating a primary node election process with one or more other VIOSes within the CIOS cluster to elect a primary node from among the VIOSes within the VIOS cluster, wherein the primary node performs one or more primary node functions of the VIOS cluster.

6. The computing electronic complex of claim 1, wherein the CA_OS further provides the function of enabling the first VIOS to provide VIO connectivity to a physical fabric connecting the first VIOS to block storage that is accessible by all VIOSes within the VIOS cluster.

7. The computing electronic complex of claim 3, wherein the CA_OS comprises a cluster management utility that executes on the virtual processor resource to provide the functions of:

providing I/O redundancy within the VIOS cluster by enabling the first VIOS to perform the functions of;

receiving an I/O request from the client LPAR;

detecting that a problem exists with a fabric connection to the block storage;

autonomously propagating the I/O request to a second VIOS to which the first VIOS is connected within the VIOS cluster, wherein forwarding of the I/O request to the block storage is completed by the second VIOS; and

in response to receiving an I/O response from the second VIOS;

associating the I/O response to the I/O request that was propagated to the second VIOS; and

forwarding the I/O response to the client LPAR of the first VIOS.

8. The computing electronic complex of claim 1, wherein the CA_OS further provides the functions of generating a software stack within the VIOS to enable the VIOS to communicate with components external to the VIOS via an application programming interface.

9. The computing electronic complex (CEC) of claim 1, wherein the CA OS further provides the function of creating the VIOS cluster with the first VIOS and one or more second VIOSes located within the CEC, where each VIOS within the VIOS cluster are communicatively coupled to each other VIOS within the VIOS cluster and have access to a shared VIOS database.

10. A data processing system (DPS) comprising:

a plurality of computing electronic complexes including the computing electronic complex (CEC) of claim 1;

a distributed storage repository that is accessible from each CEC of the plurality of CECs and by each VIOS operating within each CEC;

wherein the first VIOS is physically located on a first CEC while the second VIOS is physically located on a second CEC of the DPS; and

wherein the first VIOS and the second VIOS each comprise a virtual processor executing respective CA_OSes, are both a node within the VIOS cluster, and are both cluster-aware.

18. The method of claim 1, further comprising creating the VIOS cluster with the first VIOS and one or more second VIOSes located within the CEC, where each VIOS within the VIOS cluster are communicatively coupled to each other VIOS within the VIOS cluster and have access to a shared VIOS database.

11. In a data processing system having:

a processor, a memory coupled to the processor, at least one input/output (I/O) adapter that enables connection to an external network, a virtualization management component executing within the data processing system, and a plurality of operating system (OS) partitions including a first virtual I/O server (VIOS) partition, a method comprising;

a cluster aware (CA) operating system (OS) executing on a processor resource within the first VIOS partition to perform the functions of;

registering the first VIOS with a VIOS cluster comprising at least a second VIOS, wherein the registering of the first VIOS to the VIOS cluster enables the first VIOS to receive cluster specific data to make the first VIOS aware of the VIOS cluster, wherein the first VIOS is able to communicate information with other VIOSes within the VIOS cluster.

12. The method of claim 11, wherein the registering of the first VIOS with the VIOS cluster comprises:

initializing a cluster registration module that assigns the VIOS with a unique ID and forwards VIOS registration data to a shared VIOS database (DB);

receiving, from the shared VIOS DB, cluster configuration data and status data of the VIOS cluster for local storage at the first VIOS; and

updating a local storage of the first VIOS with the received cluster configuration data and status data.

13. The method of claim 11, further comprising:

initializing an I/O emulation module of the CA_OS to enable the VIOS to provide virtual I/O (VIO) services to one or more client logical partitions (LPARs) existing within one of the OS partitions;

registering a first client LPAR with the VIOS;

assigning a unique VIO adapter to the client LPAR; and

enabling I/O operations for the client LPAR by linking the unique VIO adapter to one of the one or more physical I/O adapters.

14. The method of claim 11, further comprising:

registering the first VIOS with the VIOS DB for receipt of one or more notifications of one or more events that may occur n the VIOS cluster;

receiving events data from within the VIOS cluster; and

updating the local storage of the first VIOS with relevant cluster information for the first VIOS.

15. The method of claim 14, further comprising coordinating a primary node election process with one or more other VIOSes within the CIOS cluster to elect a primary node from among the VIOSes within the VIOS cluster, wherein the primary node performs one or more primary node functions of the VIOS cluster.

16. The method of claim 15, further comprising:

providing I/O redundancy within the VIOS cluster by enabling the first VIOS to perform the functions of;

receiving an I/O request from the client LPAR;

detecting that a problem exists with a fabric connection to the block storage;

autonomously propagating the I/O request to a second VIOS to which the first VIOS is connected within the VIOS cluster, wherein forwarding of the I/O request to the block storage is completed by the second VIOS; and

in response to receiving an I/O response from the second VIOS;

associating the I/O response to the I/O request that was propagated to the second VIOS; and

forwarding the I/O response to the client LPAR of the first VIOS.

17. The method of claim 11, further comprising:

generating a software stack within the VIOS to enable the VIOS to communicate with components external to the VIOS via an application programming interface; and

enabling the first VIOS to provide VIO connectivity to a physical fabric connecting the first VIOS to block storage that is accessible by all VIOSes within the VIOS cluster.

19. A computer program product comprising:

a computer readable storage medium; and

program code on said computer readable storage medium that when executed by a processor within a data processing system performs the functions of;

initializing, within an operating system (OS) partition of a virtualized processing environment having one or more operating system (OS) partitions, a first virtual input/output (I/O) server (VIOS);

executing a cluster aware (CA) operating system (OS) on a processor resource within the first VIOS partition, wherein the executing of the CA OS initiates registering of the first VIOS with a VIOS cluster comprising at least a second VIOS, wherein the registering of the first VIOS to the VIOS cluster enables the first VIOS to receive cluster specific data to make the first VIOS aware of the VIOS cluster, wherein the first VIOS is able to communicate information with other VIOSes within the VIOS cluster.

20. The computer program product of claim 19, said program code that performs the registering of the first VIOS with the VIOS cluster comprises code that executes to perform the following function:

initializing a cluster registration module that assigns the VIOS with a unique ID and forwards VIOS registration data to a shared VIOS database (DB);

receiving, from the shared VIOS DB, cluster configuration data and status data of the VIOS cluster for local storage at the first VIOS; and

updating a local storage of the first VIOS with the received cluster configuration data and status data.

21. The computer program product of claim 19, wherein said program code comprises code that executes to perform the functions of:

initializing an I/O emulation module of the CA_OS to enable the VIOS to provide virtual I/O (VIO) services to one or more client logical partitions (LPARs) existing within one of the OS partitions;

registering a first client LPAR with the VIOS;

assigning a unique VIO adapter to the client LPAR; and

enabling I/O operations for the client LPAR by linking the unique VIO adapter to one of the one or more physical I/O adapters.

22. The computer program product of claim 19, wherein said program code comprises code that executes to perform the functions of:

registering the first VIOS with the VIOS DB for receipt of one or more notifications of one or more events that may occur n the VIOS cluster;

receiving events data from within the VIOS cluster; and

updating the local storage of the first VIOS with relevant cluster information for the first VIOS.

23. The computer program product of claim 19, wherein said program code comprises code that executes to perform the functions of:

creating the VIOS cluster with the first VIOS and one or more second VIOSes located within the CEC, where each VIOS within the VIOS cluster are communicatively coupled to each other VIOS within the VIOS cluster and have access to a shared VIOS database; and

coordinating a primary node election process with one or more other VIOSes within the CIOS cluster to elect a primary node from among the VIOSes within the VIOS cluster, wherein the primary node performs one or more primary node functions of the VIOS cluster.

24. The computer program product of claim 19, wherein said program code comprises code that executes to perform the functions of:

providing I/O redundancy within the VIOS cluster by enabling the first VIOS to perform the functions of;

receiving an I/O request from the client LPAR;

detecting that a problem exists with a fabric connection to the block storage;

autonomously propagating the I/O request to a second VIOS to which the first VIOS is connected within the VIOS cluster, wherein forwarding of the I/O request to the block storage is completed by the second VIOS; and

in response to receiving an I/O response from the second VIOS;

associating the I/O response to the I/O request that was propagated to the second VIOS; and

forwarding the I/O response to the client LPAR of the first VIOS.

25. The computer program product of claim 19, wherein said program code comprises code that executes to perform the functions of:

generating a software stack within the first VIOS to enable the first VIOS to communicate with components external to the VIOS via an application programming interface; and

enabling the first VIOS to provide VIO connectivity to a physical fabric connecting the first VIOS to block storage that is accessible by all VIOSes within the VIOS cluster.