Architecture for implementing a virtualization environment and appliance

• US 9,772,866 B1
• Filed: 07/17/2012
• Issued: 09/26/2017
• Est. Priority Date: 07/17/2012
• Status: Active Grant

• Current Assignee: Nutanix Incorporated
• Original Assignee: Nutanix Incorporated
• Inventors: Aron, Mohit, Pandey, Dheeraj, Singh, Ajeet, Bhardwaj, Rishi, Chun, Brent
• Primary Examiner(s): Dao, Tuan
• Application Number:
  US13/551,291
• Time in Patent Office: 1,897 Days
• Field of Search: None
• 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

An improved architecture is provided which enables significant convergence of the components of a system to implement virtualization. The infrastructure is VM-aware, and permits scaled out converged storage provisioning to allow storage on a per-VM basis, while identifying I/O coming from each VM. The current approach can scale out from a few nodes to a large number of nodes. In addition, the inventive approach has ground-up integration with all types of storage, including solid-state drives. The architecture of the invention provides high availability against any type of failure, including disk or node failures. In addition, the invention provides high performance by making I/O access local, leveraging solid-state drives and employing a series of patent-pending performance optimizations.

Claims

1. A system, comprising:

an appliance for implementing a virtualization system comprising compute components and storage components, the appliance having a plurality of serverboards, in which a respective serverboard corresponds to a single node of a plurality of nodes, wherein the single node comprises a hypervisor, one or more virtual machines, a controller virtual machine and one or more local storage devices, wherein the controller virtual machine runs as a virtual machine above the hypervisor and manages storage and I/O activities for the one or more virtual machines, wherein the appliance is rack mountable and the plurality of serverboards are inserted into the appliance, wherein one or more appliances operatively coupled to one another correspond to the virtualization system;

a plurality of sets of storage units, wherein a set of storage units from the plurality of sets of storage units comprises a local storage device that is locally attached to a corresponding serverboard of a node, and the plurality of sets of storage units comprise a storage cluster that is structured from portions of local storage devices on separate nodes; and

wherein controller virtual machines, from the plurality of nodes in the virtualization system, in communication with each other form a storage pool comprising the plurality of sets of storage units.

2. The system of claim 1, wherein the appliance has a height of 2 rack units or less.

3. The system of claim 1, wherein the appliance comprises enough redundancy such that it does not have a single point of failure.

4. The system of claim 1, wherein the appliance comprises enough redundancy such that it does not have a single point of performance bottleneck.

5. The system of claim 1, in which the appliance is-interconnected with other appliances.

6. The system of claim 1, in which the sets of storage units comprise any combination of SSDs (Solid State Drives) and HDDs (Hard Disk Drives).

7. The system of claim 1, wherein iSCSI or NFS is used to communicate between the one or more user virtual machines and the controller virtual machine via a virtual switch in the hypervisor.

8. The system of claim 1, in which a set of storage units from the plurality of sets of storage units are implemented as a shared virtual disk in a virtualization environment.

9. The system of claim 8, further comprising:

a plurality of nodes, wherein the plurality of nodes implements a virtualization environment, and a node comprises a hypervisor, user virtual machines, and a storage controller implemented as a controller virtual machine;

a plurality of storage devices that are accessed by the user virtual machines and are managed by the storage controller, in which a virtual disk is formed from the plurality of storage devices and the virtual disk can be accessed by both a first user virtual machine and a second user virtual machine;

the first user virtual machine is associated with a first storage controller and the second user virtual machine is associated with a second storage controller;

the virtual disk is owned by the first storage controller; and

I/O requests that are issued by second user virtual machine are forwarded by the second storage controller to the first storage controller.

10. The system of claim 1, further comprising:

a plurality of nodes, wherein the plurality of nodes implements a virtualization environment, and a node comprises a hypervisor, user virtual machines, and a storage controller implemented as a controller virtual machine;

a plurality of storage devices that are accessed by the user virtual machines and are managed by the storage controller, in which a first and a second virtual disk (vDisk) are formed from the plurality of storage devices and the first virtual disk can be exclusively accessed by a first user virtual machine and the second virtual disk can be exclusively accessed by a second user virtual machine;

the first user virtual machine is associated with a first storage controller and the first virtual disk is owned by the first storage controller where first virtual disk I/O requests that are issued by the first user virtual machine are handled by the first storage controller; and

the second user virtual machine is associated with a second storage controller and the second virtual disk is owned by the first storage controller where second virtual disk I/O requests that are issued by the second user virtual machine are handled by the second storage controller.

11. The system of claim 1, comprising computer readable storage medium having computer executable code for handling storage in response to migration of a virtual machine in a virtualization environment, comprising:

receiving an I/O request for a virtual disk;

determining that a controller virtual machine which implements a storage controller is not registered as owner of the virtual disk;

obtaining ownership of the virtual disk; and

handling the I/O request after obtaining ownership of the virtual disk.

12. The system of claim 1, comprising computer readable storage medium having computer executable code for handling storage in response to failure of a storage controller virtual machine in a virtualization environment, comprising:

identifying a failure of a first storage controller virtual machine in a virtualization environment, wherein the storage controller virtualization machine is owner of a virtual disk;

identifying a candidate replacement owner for the virtual disk, wherein the candidate replacement owner is a second storage controller virtual machine in the virtualization environment;

transferring ownership of the virtual disk to the candidate replacement owner; and

using the second storage controller virtual machine to handle I/O requests for the virtual disk.

13. The system of claim 1, comprising a controller virtual machine that runs as a virtual machine above a hypervisor in a node of the plurality of nodes.

14. The system of claim 1, comprising a controller virtual machine that manages a virtual disk that is exposed to a user virtual machine.

15. The system of claim 14 in which the virtual disk corresponds to one or more block devices or server targets.

16. The system of claim 1 comprising a controller virtual machine that operates as a storage controller that is dedicated to a single node.

17. The system of claim 16 in which a request for storage managed by a second service virtual machine are sent to be handled by the second service virtual machine.

18. The system of claim 16 in which a new node that is added to a system corresponds to a new controller virtual machine that acts as the storage controller for the new node.

19. The system of claim 1 comprising a controller virtual machine for each of the plurality of nodes correspond to a same IP address isolated by an internal network.

20. The system of claim 1 comprising a controller virtual machine that comprises distributed metadata service module to maintain metadata for virtual disks managed by the controller virtual machine.

21. The system of claim 1 comprising a controller virtual machine that comprises a health management module to maintain consistency of metadata for virtual disks managed by the controller virtual machine.

22. The system of claim 1 comprising a controller virtual machine that comprises a distributed configuration database module to maintain configuration data for the controller virtual machine.