STORAGE DEVICE POWER FAILURE INFRASTRUCTURE

• US 20140258755A1
• Filed: 03/07/2014
• Published: 09/11/2014
• Est. Priority Date: 03/10/2013
• Status: Active Grant

• Current Assignee: Seagate Technology LLC (Seagate Technology Holdings Plc)
• Original Assignee: LSI Corporation (Broadcom, Inc.)
• Inventors: STENFORT, Ross John
• Granted Patent: US 9,383,795 B2
• Time in Patent Office: Days
• US Class Current: 713/323
• CPC Class Codes: G06F 1/30 Means for acting in the event of power-supply failure or interruption, e.g. power-supply fluctuations for resetting only G06F1/24

Abstract

A power fail protection system wherein pluralities of individual energy storage components are electrically connected to one or more SSD drives during a power failure though a power switch matrix. Typically an individual high-energy supply will be connected to one SSD drive during a power failure. The power fail protection system may also test the transient energy response of individual energy storage components, or include an imminent power fail warning connected directly to an SSD drive interface. Some embodiments further provide for identifying, reporting, and replacing weak energy storage components. High-energy supplies may be modular, or hot swappable.

Claims

1. A system, comprising:

one or more solid state drives;

a plurality of energy storage components;

switching circuitry selectively connecting the one or more solid state drives to at least some of the plurality of energy storage components; and

wherein in the presence of a least an unavailable one of the plurality of energy storage components, via the switching circuitry, available ones of the plurality of energy storage components are enabled during a power failure to provide short-term power sufficient to enable data hardening on the one or more solid state drives.

2. The system of 1, wherein one or more of the energy storage components comprise capacitors implemented as hot-swappable removable modules comprising functional status visual indicators.

3. The system of 2, wherein the capacitors comprise tantalum capacitors.

4. The system of 2, wherein the capacitors comprise supercapacitors.

5. The system of 1, further comprising:

load discharge test circuitry enabled to perform a periodic determination of the functional status of each of the plurality of energy storage components by observing a discharge performance during a respective periodic connection of each of the plurality of energy storage components to a predetermined test load; and

wherein I/O operation performance, of the one or more solid state drives, is independent of the periodic determination without compromising the data hardening ability.

6. The system of 5, wherein the predetermined test load comprises a size chosen to implement the periodic determining faster than if the size were chosen to avoid stress of the plurality of energy storage components.

7. The system of 1, wherein the plurality of energy storage components are located within the system separately from the one or more solid state drives and to avoid heat, and the one or more solid state drives are enabled to be smaller without compromising the data hardening ability.

8. The system of 7, wherein the data hardening ability is a latent feature of the one or more solid state drives, the feature is enabled by the presence of the plurality of energy storage components, and the plurality of energy storage components are enabled to be provided as a subsequent field upgrade with respect to an earlier deployment of the plurality of storage drives.

9. The system of 1, wherein a count of the plurality of energy storage components is in excess number relative to a count of the one or more solid state drives and during the power failure each of the one or more solid state drives is connected to a respective one of the available ones of the plurality of energy storage components.

10. The system of 1, wherein the one or more solid state drives comprise a plurality of solid state drives and a count of the plurality of solid state drives is in excess number relative to a count of the available ones of the plurality of energy storage components, an excess in capacity of the available ones of the plurality of energy storage components exists relative to the needs of the one or more solid state drives, and during the power failure at least some of the available ones of the energy storage components are connected to two or more of the plurality of solid state drives.

11. The system of 1, wherein the plurality of energy storage components are collectively implemented in a common housing compatible with a selected drive form factor, the one or more solid state drives are respectively compatible with the selected drive form factor, and the common housing of energy storage components and the one or more solid state drives are removably mountable in respective drive bays of a multiple-drive-bay enclosure.

12. The system of 11, wherein the data hardening ability is a latent feature of the one or more solid state drives and the feature is enabled by later providing the plurality of energy storage components, comprised in the common housing, via a field upgrade where the

13. A method, comprising:

periodically evaluating each a plurality of energy storage components with respect to an ability to maintain a minimum voltage under a test load over a test time;

dynamically connecting, in view of a loss of a nominal power supply and the periodically evaluating, evaluated-as-good ones of the plurality of energy storage components to a plurality of storage drives; and

data hardening by each of the plurality of storage drives, subsequent to the nominal power supply loss, using reserve power supplied from the dynamically connected ones of the plurality of energy storage components.

14. The method of 13, wherein the plurality of storage drives comprise one or more solid state drives and the one or more of the energy storage components comprise supercapacitors implemented as hot-swappable removable modules comprising functional status visual indicators.

15. The method of 14, wherein the periodically evaluating is via load discharge test circuitry, the dynamically connecting is via a power switch matrix, the data hardening comprises writing data buffered in volatile storage to non-volatile storage, and in each of the solid state drives the data hardening is managed by a solid state drive controller.

16. The method of 13, wherein an excess number of a count of the plurality of energy storage components, relative to a count of the plurality of storage drives, enables more frequent testing of the energy storage components without compromising I/O operation performance and without compromising the data hardening ability.

17. The method of 13, wherein the data hardening ability is a latent feature of the plurality of storage drives, the feature is enabled by the presence of the plurality of energy storage components, and the plurality of energy storage components are enabled to be provided via a field upgrade at a time after an initial deployment of the plurality of storage drives.

18. The method of 13, further comprising:

generating at least one power status signal in accordance with predetermined criteria;

communicating the at least one power status signal to at least some of the plurality of storage drives via a pin on a standard computer interface;

operating the plurality of storage drives in an operating mode determined based at least in part on changes in the power status signal; and

wherein the functionality of the pin is configured during system initialization for at least some of the plurality of storage drives.

19. The method of 18, wherein the at least one power status signal is an imminent-power-fail signal based at least in part on the status of the nominal power supply and the operating mode comprises performing the data hardening in response to the imminent-power-fail signal indicating the loss of the nominal power supply.

20. The method of 18, wherein the at least one power status signal is a reserve power ready signal based at least in part on the ready status of at least one of the plurality of energy source components and the operating mode comprises deferring one or more operations in response to reserve power ready signal indicating a lack of readiness.

21. The method of 13, further comprising:

hot-swapping out evaluated as bad ones of the plurality of energy storage components.

22. The method of 13, further comprising:

further generating the power status signal via the comparator circuitry based at least in part on monitoring a user operated power-off switch;

wherein in response to assertion of the user operated power-off switch the power status signal communicates a power loss event in advance of a shutdown of the nominal power supply; and

wherein the data hardening occurs without reliance on the energy storage components.