# US 8,788,224 B2 · Virtual machine placement for minimizing total energy cost in a datacenter

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

- Number: US 8,788,224 B2
- Document ID: US8788224B2
- Status: Granted
- Kind: B2
- Application number: 13/584,232
- Filed: 2012-08-13
- Published: 2012-12-06
- Granted: 2014-07-22
- Earliest claimed: 2010-03-24
- Assignees: International Business Machines Corporation
- Applicants: Gargi B. Dasgupta; Pradipta De; Tapan K. Nayak; Akshat Verma
- Inventors: Gargi B. Dasgupta; Pradipta De; Tapan K. Nayak; Akshat Verma
- Web page: https://patentgrep.com/patent/US8788224B2

## CPC

- G06F 30/28: using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
- G06F 30/00: Computer-aided design [CAD]
- G06F 2119/06: Power analysis or power optimisation
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Techniques for virtual machine placement in a datacenter are provided. The techniques include using one or more server power characteristics to determine a power cost for any given virtual machine placement in the datacenter, using a heat profile of one or more datacenter components in relation to one or more cooling resources in the datacenter to determine cooling cost for any given virtual machine placement in the datacenter, and using the power cost and cooling cost to apply integrated energy cost minimization at one or more levels of hierarchy in the datacenter to determine placement of a virtual machine in the datacenter such that the integrated energy cost of the datacenter is minimized.

## Claims

### Claim 1

A computer-implemented method for virtual machine placement in a datacenter, wherein the method comprises:
using one or more server power characteristics to determine a power cost for any given virtual machine placement in the datacenter;
using a heat profile of one or more datacenter components in relation to one or more cooling resources in the datacenter to determine cooling cost for any given virtual machine placement in the datacenter; and
using the power cost and cooling cost to apply integrated energy cost minimization at one or more levels of hierarchy in the datacenter to determine placement of a virtual machine in the datacenter such that the integrated energy cost of the datacenter is minimized via:
calibrating in a bottom-up manner by creating a function of energy cost and temperature per unit of workload allocated at each level in the hierarchy; and
implementing a top-down workload placement decision-making function.

### Claim 2 (depends on 1)

The method of claim 1, wherein the one or more server power characteristics comprise processing capacity of the server.

### Claim 3 (depends on 1)

The method of claim 1, wherein the one or more datacenter components comprise at least one of a server, a rack, a row of racks and a datacenter room.

### Claim 4 (depends on 1)

The method of claim 1, wherein the one or more levels of hierarchy in the datacenter comprise at least one of one or more individual servers, a rack, a group of one or more racks, and one or more datacenter rooms.

### Claim 5 (depends on 1)

The method of claim 1, wherein the integrated energy cost equals the sum of the power cost and the cooling cost.

### Claim 6 (depends on 1)

The method of claim 1, wherein the power cost is given by a power consumed versus capacity curve for a server.

### Claim 7 (depends on 6)

The method of claim 6, wherein the power consumed versus capacity curve is built using at least one of one or more actual measurements during a calibration phase and one or more server specifications.

### Claim 8 (depends on 1)

The method of claim 1, wherein the cooling cost of is determined by a temperature versus capacity curve for each of the one or more datacenter components.

### Claim 9 (depends on 8)

The method of claim 8, wherein the temperature versus capacity curve is built for all components of the datacenter during a calibration phase using a computation fluid dynamics modeling tool that uses computation fluid dynamics to calculate temperature of a datacenter component due to heat generated by equipment, datacenter layout and one or more air flow mechanisms.

### Claim 10 (depends on 1)

The method of claim 1, wherein calibrating in a bottom-up manner comprises calibrating at a level L using a calibration at a lower level L−1 and an aggregation function that captures an impact of interference between elements at the level L−1.

### Claim 11 (depends on 10)

The method of claim 10, wherein the calibration at level L for a total workload capacity C is built using a placement on level L−1 that minimizes total energy cost at level L−1 and aggregation cost.

### Claim 12 (depends on 1)

The method of claim 1, wherein the top-down workload placement uses the bottom-up calibration to identify an optimal element at a lower level L−1 to place a workload.