Systems and methods for handling sudden power failures in solid state drives

Patent metadata

Field	Value
Publication number	US-9990023-B2
Application number	US-201615219984-A
Country	US
Kind code	B2
Filing date	Jul 26, 2016
Priority date	Mar 28, 2016
Publication date	Jun 5, 2018
Grant date	Jun 5, 2018

Abstract

Systems and methods for handling sudden power failures in non-volatile memory devices such as solid state drives are provided by having the non-volatile memory device boot up in a low power write mode, which limits substantially all programming operations to a single level cell (SLC) mode, as opposed to a normal mode in which the programming operations can be performed in a multi-level cell (MLC) mode. Thus, if the system experiences a sudden power failure when it is being powered solely by AC derived power and the battery is below a level sufficient for powering the device while it is programming in the SLC mode, data integrity will be preserved because the programming operation was being performed in SLC mode. The non-volatile memory device may be permitted to exit out the low power write mode into the normal mode when the charge level of the battery is sufficient for powering the system.

First claim

What is claimed is:

1. A system, comprising
an AC derived power port;
a battery system comprising a battery having a charge level;
a non-volatile memory (NVM) storage device operative to receive power from at least one of the battery system and the AC derived power port, the NVM storage device comprising a plurality of memory cells that store data according to one of a single level cell (SLC) mode and a multi-level cell (MLC) mode;
control circuitry operative to:
boot the NVM storage device in a low power write mode (LPWM) that forces all program operations to be performed in the SLC mode; and
instruct the NVM storage device to exit out of the LPWM into a normal mode when the charge level of the battery is at the charge level above a minimum predetermined threshold; and

power management circuitry coupled to the battery system, NVM storage device, control circuitry, and the AC derived power port, wherein when the NVM storage device is in the LPWM, the control circuitry instructs the power management circuitry to set its delay timing parameter to a LPWM SPF time period, and wherein when the NVM storage device is in the normal mode, the control circuitry instructs the power management circuitry to set its delay timing parameter to a normal mode SPF time period, where the normal mode SPF time period is at least one order of magnitude longer than the LPWM SPF time period.

2. The system of claim 1 , wherein in the LPWM, garbage collection program operations are the only program operations performed in the MLC mode.

3. The system of claim 1 , wherein the LPWM preserves data integrity of data stored in at least a portion of the plurality of memory cells being programmed during a sudden power failure (SPF) event.

4. The system of claim 3 , wherein the SPF event occurs when the charge level is at the charge below the predetermined threshold and the AC derived power port abruptly ceases to supply power to the NVM storage device.

5. The system of claim 1 , wherein in the normal mode, program operations are performed in the MLC mode.

6. The system of claim 1 , wherein the delay timing parameter defines a maximum delay period between reset and write protect.

7. The system of claim 1 , further comprising power failure detect circuitry coupled to the NVM storage device, the battery system, and the AC derived power port, the power failure detect circuitry operative to provide a SPF notification to the NVM storage device, wherein the NVM storage device, in response to receiving the SPF notification during a programming operation and while in the LPWM, is operative to:
cease issuing program commands; and
terminate all in-flight program operations.

8. The system of claim 1 , wherein the system is a computer and the NVM storage device is a solid state drive (SSD).

9. A method for data storage, comprising:
booting a solid state drive (SSD) in a low power write mode (LPWM), wherein the LPWM preserves data integrity of data stored in at least a portion of the SSD being programmed during a sudden power failure (SPF) event;
determining if a charge level of a battery is above a threshold;
if the charge level is below the threshold, enforcing the LPWM such that all program operations are performed in a single-level cell (SLC) mode;
if the charge level is above the threshold, exiting out of the LPWM such that the SSD operates in a normal mode, wherein in the normal mode, program operations are performed in a multi-level cell (MLC) mode; and
receiving a SPF notification while programming data in the LPWM;
ceasing further issuance of program commands in response to the received SPF notification; and
terminating all in-flight program operations in response to the received SPF notification.

10. The method of claim 9 , further comprising supplying power from an AC derived power port to the SSD when the charge level is less than the threshold, wherein the SPF event occurs when the charge level is at the charge below the threshold and the AC derived power port abruptly ceases to supply power to the SSD.

11. The method of claim 9 , wherein the enforcing comprises enabling garbage collection program operations to be performed in the MLC mode.

12. The method of claim 9 , wherein the enforcing comprises enabling read and erase operations to be performed according to the normal mode.

13. The method of claim 9 , further comprising preventing the SSD from returning to the LPWM after the SSD had previously exited out of the LPWM mode, wherein the preventing persists until the SSD cold boots in the LPWM mode.

14. The method of claim 9 , further comprising changing a delay timing parameter in power management circuitry to comply with one of the LPWM and the normal mode, wherein the delay timing parameter defines a time between assertions of write protect and reset of the SSD.

15. A method for handling sudden power failures in an apparatus comprising a battery system, an AC derived power port, a solid state drive (SSD), and control circuitry, the method comprising:
cold booting the SSD in a low power write mode (LPWM), wherein the LPWM preserves data integrity of data stored in at least a portion of the SSD being programmed during a sudden power failure (SPF) event that occurs when the battery system is unable to power the apparatus and the apparatus is being powered by power received on the AC derived power port, but the power being received on the AC derived power port is abruptly cut off before the battery is able to power the apparatus;
programming data in a single level cell mode while the SSD is in the LPWM;
transitioning the SSD from the LPWM to a normal mode when the battery is able to power the apparatus, wherein in the normal mode, the SSD programs data in a multi-level cell mode; and
changing a delay timing parameter in power management circuitry to comply with one of the LPWM and the normal mode, wherein the delay timing parameter defines a time between assertions of write protect and reset of the SSD, wherein the delay timing parameter associated with the LPWM mode is on order of magnitude less than the delay timing parameter associated with the normal mode.

Assignees

• Apple Inc

Inventors

• Paley Alexander
• Vogan Andrew W
• Sandel Eran
• Mouler Lior
• Erez Liran
• Byom Matthew J
• Ashraf Muhammad N
• Guy Roman

Classifications

• G06F11/3058

  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 title

• G06F1/3275 Primary

  Power saving in memory, e.g. RAM, cache · CPC title

• G06F11/3034

  where the computing system component is a storage system, e.g. DASD based or network based (digital input from or digital output to record carriers G06F3/06; digital recording or reproducing G11B20/18; for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS], H04L67/1097) · CPC title

• G06F11/004

  Error avoidance (G06F11/07 and subgroups take precedence) · CPC title

• G06F1/3287

  by switching off individual functional units in the computer system · CPC title