# US 2014/0157269 A1 · Common contiguous memory region optimized long distance virtual machine migration

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2014/0157269 A1
- Document ID: US20140157269A1
- Status: Published application
- Kind: A1
- Application number: 13/788,428
- Filed: 2013-03-07
- Published: 2014-06-05
- Earliest claimed: 2012-11-30
- Assignees: International Business Machines Corporation
- Applicants: International Business Machines Corporation
- Inventors: Eli M. Dow; James P. Gilchrest; Steven K. Schmidt; Jessie Yu
- Web page: https://patentgrep.com/patent/US20140157269A1

## CPC

- G06F 9/45558: {Hypervisor-specific management and integration aspects}
- G06F 2009/4557: {Distribution of virtual machine instances; Migration and load balancing}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Embodiments relate to migrating a virtual machine (VM) between work groups. A first machine in a first work group generates a migration request that identifies one or more contiguous memory regions used by the VM for operation. The first machine transmits the migration request to a plurality of work groups. The first machine receives a score from each of the plurality of work groups indicating a number of contiguous memory regions identified in the migration request that are available at the respective work group. A second work group in the plurality of work groups is identified as a destination for the VM based on the scores and a distance of the second work group from the first machine. A migration of the VM from the first machine to a second machine in the second work group is caused based on the identification of the second work group as the destination.

## Claims

### Claim 1

A method for migrating a virtual machine (VM) between work groups, the method comprising:
generating, by a first machine in a first work group, a migration request that identifies one or more contiguous memory regions used by the VM for operation;
transmitting, by the first machine, the migration request to a plurality of work groups;
receiving, by the first machine, a score from each of the plurality of work groups indicating a number of contiguous memory regions identified in the migration request that are available at the respective work group;
identifying a second work group in the plurality of work groups as a destination for the VM based on the scores and a distance of the second work group from the first machine; and
causing a migration of the VM from the first machine to a second machine in the second work group based on the identification of the second work group as the destination.

### Claim 2 (depends on 1)

The method of claim 1, wherein the first machine is a first edge machine and the second machine is a second edge machine.

### Claim 3 (depends on 2)

The method of claim 2, wherein the causing of a migration from the first machine to the second machine is based on the second edge machine determining that the second machine has available the most contiguous memory identified in the migration request in the second work group.

### Claim 4 (depends on 1)

The method of claim 1, wherein the transmitting is to a second edge machine included in the second work group, and a first edge machine in the first work group receives the score for the second work group from the second edge machine.

### Claim 5 (depends on 1)

The method of claim 1, further comprising:
identifying the second work group as the destination for the VM based on a weighting of each of the scores relative to the distances of each work group included in the plurality of work groups.

### Claim 6 (depends on 1)

The method of claim 1, further comprising:
identifying the second work group as the destination for the VM based on an activity level associated with the VM.

### Claim 7 (depends on 1)

The method of claim 1, further comprising:
using at least one of a virtual private network (VPN) and a tunnel to migrate the VM from the first machine to the second machine in order to minimize a number of hops.