DATA CENTER MANAGEMENT UNIT WITH IMPROVED DISASTER PREVENTION AND RECOVERY

• US 20110320834A1
• Filed: 08/12/2010
• Published: 12/29/2011
• Est. Priority Date: 12/03/2009
• Status: Active Grant

• Current Assignee: GIG Technology NV
• Original Assignee: Racktivity NV (GIG Technology NV)
• Inventors: Ingels, Wilbert, Vinken, Niko
• Granted Patent: US 8,712,558 B2
• Time in Patent Office: Days
• US Class Current: 713/310
• CPC Class Codes: G06F 1/266 Arrangements to supply power to external peripherals either directly from the computer or under computer control, e.g. supply of power through the communication port, computer controlled power-strips
• CPC Class Codes: G06F 1/28 Supervision thereof, e.g. detecting power-supply failure by out of limits supervision
• CPC Class Codes: G06F 1/305 in the event of power-supply fluctuations

Abstract

A data center management unit for managing and controlling power distribution to computers in a data center includes a power inlet, a plurality of power outlets for providing power to respective ones of the computers, and a processor. In addition thereto, the data center management unit includes a memory for storing logged parameters or events. The processor in the data center management unit takes measures for disaster prevention and/or disaster recovery based on the logged parameters or events. The measures for disaster prevention or disaster recovery comprise generating alert messages, de-activating one or more power outlets, activating one or more power outlets and/or re-activating a power outlet that has been de-activated earlier.

Claims

1. A data center management unit for managing and controlling power distribution to computers in a data center, said data center management unit comprising a power inlet, a plurality of power outlets for providing power to respective ones of said computers, and a processor,

wherein said data center management unit further comprises memory means for storing one or more logged parameters or events, and in that said processor is adapted to take measures for disaster prevention and/or disaster recovery based on said logged parameters or events, said measures for disaster prevention or disaster recovery comprising one or more of the following:

generating alert messages;

de-activating one or more power outlets of said plurality of power outlets;

activating one or more power outlets of said plurality of power outlets; and

re-activating a power outlet that has been de-activated earlier.

2. A data center management unit according to claim 1, wherein said processor is adapted to analyse changes in time of values of said logged parameters or events.

3. A data center management unit according to claim 2, wherein analysing said changes in time comprises comparing short term values, logged for a short term between zero and fifteen minutes, and long term values, logged for a long term being 1 day or more, of said logged parameters or events.

4. A data center management unit according to claim 1, wherein said data center management unit comprises means for detecting micro-interruptions in power, measuring durations of said micro-interruptions, and measuring phase shifts in relation to such micro-interruptions.

5. A data center management unit according to claim 1, wherein said data center management unit further comprises:

power metering circuitry for measuring overall power consumption at said power inlet; and

wherein

said processor is adapted to apply a first priority scheme for de-activating power outlets of said plurality of power outlets in a first priority order whenever said overall power consumption exceeds an overall threshold.

6. A data center management unit according to claim 1,

wherein said first priority scheme is based on comparison of long term values, logged for a long term being 1 day or more, and short term values, logged for a short term between zero and fifteen minutes, of said logged parameter values or events;

and wherein

in case of equal priority based on said comparison a power outlet which is more recently activated is prioritized for de-activation.

7. A data center management unit according to claim 1,

wherein said processor is adapted to apply a second priority scheme for re-activating power outlets out of said plurality of power outlets in a second priority order after a power outage.

8. A data center management unit according to claim 7,

wherein said second priority scheme is based on comparison of long term values, logged for a long term being 1 day or more, and short term values, logged for a short term between zero and fifteen minutes, of said logged parameters or events.

9. A data center management unit according to claim 1,

wherein said processor is adapted to de-activate a power outlet of said plurality of power outlets whenever a power measurement for said power outlet exceeds a predetermined per-outlet threshold.

10. A data center management unit according to claim 1,

said data center management unit further comprising memory means for storing one or more parameter values during start-up of a computer out of said computers.

11. A data center management unit according to claim 10,

wherein said processor is adapted to determine sequence and timing of re-starting said computers after a power disaster based on said parameter values measured during start-up of said computers.

12. A data center management unit according to claim 10,

wherein said processor is adapted to instruct deactivation of a computer or alerting when parameter values measured during operation of said computer approach said parameter values measured during start-up of said computer within predetermined limits.

13. A data center management unit according to claim 1,

said data center management unit further comprising memory means for storing the frequency spectrum of current and voltage measured at a power outlet of said plurality of power outlets.

14. A data center management unit according to claim 1,

wherein said processor is adapted to analyse harmonics in said frequency spectrum for threat and failure prediction.

15. A data center management unit according to claim 1,

wherein said logged parameters or events comprise one or more of the following:

temperature in said data center;

humidity in said data center;

airflow in said data center;

current measured at a power outlet of said plurality of power outlets;

total harmonic distortion or THD;

power factor of a plurality of said power outlets;

voltage measured at said power inlet;

energy consumed by a power outlet of said plurality of power outlets;

power consumed by a power outlet of said plurality of power outlets;

leakage current at a power outlet of said plurality of power outlets;

short circuit events.

16. A method for managing and controlling power distribution towards computers in a data center via a data center management unit comprising a power inlet, a plurality of power outlets and a processor,

wherein said method comprises:

storing one or more logged parameters or events; and

taking measures for disaster prevention and/or disaster recovery based on said logged parameters or events, said measures for disaster prevention or disaster recovery comprising one or more of the following;

generating alert messages;

de-activating one or more power outlets of said plurality of power outlets;

activating one or more power outlets of said plurality of power outlets; and

re-activating a power outlet that has been de-activated earlier.