Administering virtual machines in a distributed computing environment

• US 9,503,515 B2
• Filed: 04/25/2014
• Issued: 11/22/2016
• Est. Priority Date: 04/24/2014
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Archer, Charles J., Blocksome, Michael A., Carey, James E., Sanders, Philip J.
• Primary Examiner(s): KESSLER, GREGORY AARON
• Application Number:
  US14/261,874
• Publication Number:
  US 20150309824A1
• Time in Patent Office: 942 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 2009/45595 Network integration; Enabling network access in virtual machine instances
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 67/131 Protocols for games, networked simulations or virtual reality

Abstract

In a distributed computing environment that includes hosts that execute a VMM, where each VMM supports execution of one or more VMs, administering VMs may include: assigning, by a VMM manager, the VMMs of the distributed computing environment to a logical tree topology, including assigning one of the VMMs as a root VMM of the tree topology; and executing, amongst the VMMs of the tree topology, a broadcast operation, including: pausing, by the root VMM, execution of one or more VMs supported by the root VMM; sending, by the root VMM, to other VMMs in the tree topology, a message indicating a pending transfer of the paused VMs; and transferring the paused VMs from the root VMM to the other VMMs.

Claims

1. A method of administering a plurality of virtual machines (‘

VMs’

) in a distributed computing environment, the distributed computing environment comprising a plurality of hosts, one or more of the hosts executing a virtual machine monitor (‘

VMM’

), each VMM supporting execution of one or more VMs, the method comprising;

assigning, by a VMM manager, the VMMs of the distributed computing environment to a logical tree topology, including assigning one of the VMMs as a root VMM of the tree topology; and

executing, amongst the VMMs of the tree topology, an allgather operation, including;

sending, by the root VMM, to a plurality of other VMMs in the tree topology, a request to retrieve VMs supported by the plurality of other VMMs;

pausing, by each of the plurality of other VMMs, a VM supported by the VMM;

providing, by each of the plurality of other VMMs as a response to the root VMM'"'"'s request, each paused VM supported by the plurality of other VMMs; and

broadcasting, by the root VMM, a same set of the received VMs to each of the plurality of other VMMs.

2. The method of claim 1, further comprising:

after the receipt of the set of VMs broadcasted by the root VMM, executing, by all VMMs, the corresponding set of VMs received.

3. The method of claim 1, further comprising:

after receiving the set of VMs broadcasted by the root VMM, storing, by all VMMs, each of the set of VMs without executing the set of VMs.

4. The method of claim 1, further comprising:

after broadcasting the set of VMs, executing, only by the root VMM, each of the set of VMs.

5. The method of claim 1, further comprising:

after receiving the set of VMs, executing, only by VMMs other than the root VMM, each of the set of VMs.

6. The method of claim 1, wherein executing the allgather operation further comprises executing the allgather operation periodically at predefined intervals without user intervention.

7. The method of claim 1, wherein executing the allgather operation further comprises executing the allgather operation on-demand responsive to a user request.

8. The method of claim 1, wherein executing the allgather operation further comprises executing the allgather operation, automatically without user intervention, in dependence upon an execution policy and responsive to an event specified in the execution policy.

9. The method of claim 1, wherein the distributed computing environment comprises a parallel computer, the parallel computer comprising a plurality of compute nodes, each of the compute nodes comprising one of the plurality of hosts and executing at least one of the VMMs, each compute node further comprising a plurality of communications adapters, each communications adapter configured to couple the compute node to other compute nodes for data communications and to one of a plurality of data communications networks.

10. The method of claim 9, wherein one of the plurality of data communications networks comprises a multi-dimensional, point-to-point network.

11. The method of claim 9, wherein one of the plurality of data communications networks comprises a global combining network.

12. The method of claim 1 further comprising:

after receiving the set of VMs, executing, only by VMMs other than the root VMM, each of the set of VMs.

13. The method of claim 1, wherein broadcasting, by the root VMM, the same set of the received VMs to each of the plurality of other VMMs further including:

sending, by the root VMM, a message indicating a pending transfer of the set of VMs to each of the plurality of other VMMs; and

transferring, by the root VMM, the set of VMs from the root VMM to each of the plurality of other VMMs.