# US 7,613,576 B2 · Using EMI signals to facilitate proactive fault monitoring in computer systems

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

- Number: US 7,613,576 B2
- Document ID: US7613576B2
- Status: Granted
- Kind: B2
- Application number: 11/787,003
- Filed: 2007-04-12
- Published: 2008-10-16
- Granted: 2009-11-03
- Earliest claimed: 2007-04-12
- Assignees: Sun Microsystems, Inc.
- Inventors: Kenny C. Gross; Aleksey M. Urmanov; Ramakrishna C. Dhanekula; Steven F. Zwinger
- Web page: https://patentgrep.com/patent/US7613576B2

## CPC

- G06F 11/00: Error detection; Error correction; Monitoring (error detection, correction or monitoring in information storage based on relative movement between record carrier and transducer G11B20/18; monitoring, i.e. supervising the progress of recording or reproducing G11B27/36; in static stores G11C29/00)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A system that monitors electromagnetic interference (EMI) signals to facilitate proactive fault monitoring in a computer system is presented. During operation, the system receives EMI signals from one or more antennas located in close proximity to the computer system. The system then analyzes the received signals to proactively detect anomalies during operation of the computer system.

## Claims

### Claim 1

A method for monitoring electromagnetic interference (EMI) signals to facilitate proactive fault monitoring in a computer system, comprising:
receiving at an analysis mechanism EMI signals from one or more antennas located in close proximity to the computer system;
using a Fast Fourier Transform (FFT) to convert the received signals to the frequency domain;
converting the output of the FFT into time series frequency signals for different frequencies;
determining correlations between the time series frequency signals and other system telemetry signals to identify time series frequency signals which are highly correlated with the other telemetry signals;
constructing a model during a training phase by determining correlations between a set of signals, wherein the set of signals includes the identified time series frequency signals and the other telemetry signals; and
comparing the time series frequency signals with corresponding estimated time series signals produced by the model to detect anomalies during a monitoring phase.

### Claim 2 (depends on 1)

The method of claim 1, wherein the determining the correlations involves using a non-linear, non-parametric regression technique.

### Claim 3 (depends on 2)

The method of claim 2, wherein the non-linear, non-parametric regression technique can include a multivariate state estimation technique.

### Claim 4 (depends on 1)

The method of claim 1, wherein comparing the time series frequency signals with corresponding estimated time series signals during the monitoring phase involves using sequential detection techniques to detect anomalies.

### Claim 5 (depends on 4)

The method of claim 4, wherein the sequential detection techniques include the Sequential Probability Ratio Test (SPRT).

### Claim 6 (depends on 1)

The method of claim 1, wherein the other system telemetry signals can include:
internal performance signals containing parameters maintained by software within the computer system;
physical performance signals measured through sensors located in proximity to the computer system; and
canary performance signals containing parameters associated with synthetic user transactions periodically generated for performance measuring purposes.

### Claim 7

A computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method for monitoring electromagnetic interference (EMI) signals to facilitate proactive fault monitoring in a computer system, the method comprising:
receiving EMI signals from one or more antennas located in close proximity to the computer system;
using a Fast Fourier Transform (FFT) to convert the received signals to the frequency domain;
converting the output of the FFT into time series frequency signals for different frequencies;
determining correlations between the time series frequency signals and other system telemetry signals to identify time series frequency signals which are highly correlated with the other telemetry signals;
constructing a model during a training phase by determining correlations between a set of signals, wherein the set of signals includes the identified time series frequency signals and the other telemetry signals; and
comparing the time series frequency signals with corresponding estimated time series signals produced by the model to detect anomalies during a monitoring phase.

### Claim 8 (depends on 7)

The computer-readable storage medium of claim 7, wherein the determining the correlations involves using a non-linear, non-parametric regression technique.

### Claim 9 (depends on 8)

The computer-readable storage medium of claim 8, wherein the non-linear, non-parametric regression technique can include a multivariate state estimation technique.

### Claim 10 (depends on 7)

The computer-readable storage medium of claim 7, wherein comparing the time series frequency signals with corresponding estimated time series signals during the monitoring phase involves using sequential detection techniques to detect anomalies.

### Claim 11 (depends on 10)

The computer-readable storage medium of claim 10, wherein the sequential detection techniques include the Sequential Probability Ratio Test (SPRT).

### Claim 12 (depends on 7)

The computer-readable storage medium of claim 7, wherein the other system telemetry signals can include:
internal performance signals containing parameters maintained by software within the computer system;
physical performance signals measured through sensors located in proximity to the computer system; and
canary performance signals containing parameters associated with synthetic user transactions periodically generated for performance measuring purposes.

### Claim 13

An apparatus that monitors electromagnetic interference (EMI) signals to facilitate proactive fault monitoring in a computer system, comprising:
one or more antennas located in close proximity to the computer system wherein the one or more antennas are configured to receive EMI signals from the computer system; and
an analysis mechanism coupled to the one or more antennas, wherein the analysis mechanism is configured to
use a Fast Fourier Transform (FFT) to convert the received signals to the frequency domain;
convert the output of the FFT into time series frequency signals for different frequencies;
determine correlations between the time series frequency signals and the other system telemetry signals to identify time series frequency signals which are highly correlated with the other telemetry signals;
construct the model by determining correlations between a set of signals, wherein the set of signals includes the identified time series frequency signals and the other telemetry signals; and
compare the time series frequency signals with corresponding estimated time series signals produced by the model to detect anomalies during a monitoring phase.