Data center monitoring based on electromagnetic wave detection

• US 10,234,492 B2
• Filed: 08/31/2016
• Issued: 03/19/2019
• Est. Priority Date: 08/31/2016
• Status: Active Grant

• Current Assignee: CA, Inc. (d/b/a CA Technologies) (Broadcom, Inc.)
• Original Assignee: CA, Inc. (d/b/a CA Technologies) (Broadcom, Inc.)
• Inventors: Gupta, Smrati, Ferrin, Thomas Howard, Muntes-Mulero, Victor
• Primary Examiner(s): Desta, Elias
• Application Number:
  US15/252,985
• Publication Number:
  US 20180060203A1
• Time in Patent Office: 930 Days
• Field of Search: 702182
• CPC Class Codes: G01R 29/0878 Sensors; antennas; probes; detectors wave guide measuring sections G01R1/24
• CPC Class Codes: G01R 29/0892 Details related to signal analysis or treatment; presenting results, e.g. displays; measuring specific signal features other than field strength, e.g. polarisation, field modes, phase, envelope, maximum value
• CPC Class Codes: G06F 11/3058 Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations thermal management in cooling arrangements of a computing system G06F1/206
• CPC Class Codes: G06F 11/3442 for planning or managing the needed capacity

Abstract

A non-intrusive system which monitors a data center by detecting electromagnetic waves alleviates a resource burden on components and is inexpensive to deploy and scale within a data center. The system detects waves using omnidirectional antennas positioned throughout the data center, thus alleviating the need to physically attach directional antennas to components. The system performs a learning phase wherein representations of detected waves are mapped to occurrence of events within the data center. Once the learning phase is complete, operation of existing network monitoring tools, such as agents and probes, may cease, and the system may begin monitoring for events based on the detected waves. The system may also analyze wave data prior to the occurrence of events to identify event prediction indicators, e.g. distinctive wave values or patterns, which may be used to predict the occurrence of an event.

Claims

1. A method comprising:

detecting first electromagnetic waves radiated by a plurality of devices in a data center;

based on receiving an indication of a first event occurring at a first device of the plurality of devices,

capturing first measurements from the first electromagnetic waves based, at least in part, on a time of the first event;

generating a first representation of the first electromagnetic waves based, at least in part, on the first measurements of the first electromagnetic waves; and

mapping the first representation to an identifier for the first event.

2. The method of claim 1 further comprising:

based on detecting second electromagnetic waves radiated by the plurality of devices,

sampling the second electromagnetic waves to generate a set of samples;

generating a set of representations based, at least in part, on the set of samples;

comparing each of the set of representations to the first representation of the first electromagnetic waves; and

based on detecting a match between a second representation of the set of representations and the first representation, indicating that the first event has occurred.

3. The method of claim 2, wherein mapping the first representation to the identifier for the first event comprises storing the first representation along with the identifier in an entry in a database, wherein comparing each of the set of representations to the first representation comprises querying the database with each of the set of representations.

4. The method of claim 2, wherein sampling the second electromagnetic waves to generate the set of samples comprises capturing samples at specified time intervals while the second electromagnetic waves are being detected, wherein a sample comprises values of the second electromagnetic waves for a period of time.

5. The method of claim 1 further comprising:

capturing second measurements from the first electromagnetic waves based, at least in part, on a time period prior to the time of the first event;

comparing the second measurements to measurements of second electromagnetic waves, wherein the second electromagnetic waves are electromagnetic waves that were detected during a normal operating period of the plurality of devices; and

identifying an event prediction indicator in the second measurements, wherein the event prediction indicator is a value in the second measurements that differs from a value in the measurements of the second electromagnetic waves.

6. The method of claim 5 further comprising:

based on detecting third electromagnetic waves radiated by the plurality of devices,

determining that a value in the third electromagnetic waves matches the event prediction indicator; and

based on determining that the value in the third electromagnetic waves matches the event prediction indicator, generating an indication that the first event may occur.

7. The method of claim 1, wherein generating the first representation of the first electromagnetic waves based, at least in part, on the first measurements of the first electromagnetic waves comprises generating a signature of the first measurements.

8. The method of claim 1, wherein detecting the first electromagnetic waves radiated by the plurality of devices in the data center comprises receiving the first electromagnetic waves from one or more omnidirectional antennas positioned throughout the data center to capture the first electromagnetic waves from the plurality of devices.

9. The method of claim 1, wherein capturing the first measurements from the first electromagnetic waves based, at least in part, on the time of the first event comprises:

tagging the first electromagnetic waves with a first timestamp corresponding to a time the first electromagnetic waves were detected;

determining the time of the first event based, at least in part, on a second timestamp associated with the indication of the first event; and

capturing the first measurements from a portion of the first electromagnetic waves that were detected at a time period comprising the time of the first event, wherein the portion of the first electromagnetic waves that were detected at the time period comprising the time of the first event is identified based, at least in part, on the first timestamp and the second timestamp.

10. One or more non-transitory machine-readable media having program code for electromagnetic wave based event detection stored therein, the program code to:

create a repository of mappings between system events of a plurality of devices in a data center and first representations of past electromagnetic waves that occurred in the data center;

generate second representations of electromagnetic waves that occur in the data center;

compare the second representations with the first representations; and

based on detection of a matching representation between the first and the second representations, generate an indication of a system event that is mapped to the matching representation in the repository.

11. The non-transitory machine-readable media of claim 10, wherein the program code to create the repository of mappings comprises program code to:

record representations of electromagnetic waves detected in the data center by a set of one or more antennas;

associate system events detected at the plurality of devices with the representations based, at least in part, on timestamps of the system events; and

create a mapping between a system event and a representation based, at least in, part, on a confidence value of an association between the system event and the electromagnetic wave exceeding a threshold.

12. The non-transitory machine-readable media of claim 10, wherein the program code to generate the second representations of electromagnetic waves that occur in the data center comprises program code to generate the second representations for electromagnetic waves detected in the data center for a sliding window of time.

13. An apparatus comprising:

a processor; and

a machine-readable medium having program code executable by the processor to cause the apparatus to,

detect first electromagnetic waves radiated by a plurality of devices in a data center;

based on receipt of an indication of a first event occurring at a first device of the plurality of devices,

capture first measurements from the first electromagnetic waves based, at least in part, on a time of the first event;

generate a first representation of the first electromagnetic waves based, at least in part, on the first measurements of the first electromagnetic waves; and

map the first representation to an identifier for the first event.

14. The apparatus of claim 13 further comprising program code executable by the processor to cause the apparatus to:

based on detection of second electromagnetic waves radiated by the plurality of devices,

sample the second electromagnetic waves to generate a set of samples;

generate a set of representations based, at least in part, on the set of samples;

compare each of the set of representations to the first representation of the first electromagnetic waves; and

based on detecting a match between a second representation of the set of representations and the first representation, indicate that the first event has occurred.

15. The apparatus of claim 14, wherein the program code executable by the processor to cause the apparatus to map the first representation to the identifier for the first event comprises program code executable by the processor to cause the apparatus to store the first representation along with the identifier in an entry in a database, wherein the program code executable by the processor to cause the apparatus to compare each of the set of representations to the first representation comprises program code executable by the processor to cause the apparatus to query the database with each of the set of representations.

16. The apparatus of claim 14, wherein the program code executable by the processor to cause the apparatus to sample the second electromagnetic waves to generate the set of samples comprises program code executable by the processor to cause the apparatus to capture samples at specified time intervals while the second electromagnetic waves are being detected, wherein a sample comprises values of the second electromagnetic waves for a period of time.

17. The apparatus of claim 13 further comprising program code executable by the processor to cause the apparatus to:

capture second measurements from the first electromagnetic waves based, at least in part, on a time period prior to the time of the first event;

compare the second measurements to measurements of second electromagnetic waves, wherein the second electromagnetic waves are electromagnetic waves that were detected during a normal operating period of the plurality of devices; and

identify an event prediction indicator in the second measurements, wherein the event prediction indicator is a value in the second measurements that differs from a value in the measurements of the second electromagnetic waves.

18. The apparatus of claim 17 further comprising program code executable by the processor to cause the apparatus to:

based on detection of third electromagnetic waves radiated by the plurality of devices,

determine that a value in the third electromagnetic waves matches the event prediction indicator; and

based on a determination that the value in the third electromagnetic waves matches the event prediction indicator, generate an indication that the first event may occur.

19. The apparatus of claim 13, wherein the program code executable by the processor to cause the apparatus to generate the first representation of the first electromagnetic waves based, at least in part, on the first measurements of the first electromagnetic waves comprises program code executable by the processor to cause the apparatus to generate a signature of the first measurements.

20. The apparatus of claim 13, wherein the program code executable by the processor to cause the apparatus to detect the first electromagnetic waves radiated by the plurality of devices in the data center comprises program code executable by the processor to cause the apparatus to receive the first electromagnetic waves from one or more omnidirectional antennas positioned throughout the data center to capture the first electromagnetic waves from the plurality of devices.