Storage subsystem with configurable buffer

• US 8,151,020 B2
• Filed: 09/24/2009
• Issued: 04/03/2012
• Est. Priority Date: 02/07/2007
• Status: Expired due to Fees

• Current Assignee: Western Digital Technologies Incorporated (Western Digital Corporation)
• Original Assignee: SiliconSystems, Inc. (Western Digital Corporation)
• Inventors: Merry, David E. Jr., Diggs, Mark S.
• Primary Examiner(s): Tsai, Henry
• Assistant Examiner(s): RHU, KRIS M
• Application Number:
  US12/566,453
• Publication Number:
  US 20100017542A1
• Time in Patent Office: 922 Days
• Field of Search: 710/56
• US Class Current: 710/56
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0656 Data buffering arrangements
• CPC Class Codes: G06F 3/0679 Non-volatile semiconductor memory device, e.g. flash memory, one time programmable memory [OTP]
• CPC Class Codes: G11B 19/044 by using a data buffer

Abstract

A storage subsystem includes a variable-size write buffer that temporarily stores write data received from a host system. The storage subsystem is capable of adjusting the size of the write buffer so as to vary both the performance (e.g., sustained write speed) of the storage subsystem and a risk of data loss. In one embodiment, the storage subsystem implements a command set that enables the host system to directly control the size of the write buffer. The storage subsystem may additionally or alternatively be capable of adjusting the size of the write buffer based on monitored operating conditions, such as the temperature, the stability/consistency of a power signal received from the host system, and/or the elapsed time since the storage subsystem was last powered up.

Claims

1. A storage subsystem, comprising:

a controller coupled to non-volatile storage, the controller capable of writing data to, and reading data from, the non-volatile storage in response to commands received by the storage subsystem from a host system; and

a buffer memory, the buffer memory capable of receiving data at a higher rate than the non-volatile storage,

wherein the controller is configured to implement a write buffer in the buffer memory to buffer write data received from the host system and transfer the write data from the write buffer to the non-volatile storage, and to automatically adjust a size of the write buffer based at least in part on one or more monitored conditions of the storage subsystem that are reflective of a likelihood of a data loss in the storage subsystem, the automatically adjusting including reducing the size of the write buffer between at least a first non-zero size value and a second non-zero size value in response to the one or more monitored conditions indicating an increased risk of data loss, so as to vary both a sustained write speed of the storage subsystem and a risk of data loss.

2. The storage subsystem of claim 1, wherein the one or more monitored conditions comprise a temperature reading of the storage subsystem obtained from a temperature sensor.

3. The storage subsystem of claim 1, wherein the storage subsystem further comprises a backup power source and the one or more monitored conditions comprise a charge level of the backup power source.

4. The storage subsystem of claim 1, wherein the one or more monitored conditions comprise at least one of:

a length of time that the storage subsystem has been powered up; and

an average length of time that the storage subsystem has remained powered up.

5. The storage subsystem of claim 1, wherein the one or more monitored conditions comprise at least one of:

a length of time since a power signal anomaly has been detected; and

a percentage of time in which a power signal from the host system has been anomalous since a last power-up.

6. The storage subsystem of claim 1, wherein the one or more monitored conditions comprise a percentage of time the storage subsystem spent in executing write operations since a last power-up.

7. The storage subsystem of claim 1, wherein the one or more monitored conditions comprise information regarding a number of sectors in which ECC errors have been detected in read operations of the non-volatile storage.

8. The storage subsystem of claim 1, wherein the storage subsystem further comprises a vibration sensor and the one or more monitored conditions comprise an output of the vibration sensor.

9. The storage subsystem of claim 1, wherein the non-volatile storage further comprises a restricted area configured to store configuration data for the controller to configure the buffer memory.

10. The storage subsystem of claim 1, in combination with the host system, wherein the host system is configured to cause the controller to vary the size of the write buffer based on a calculated probability of failure based on events logged by the host system.

11. The storage subsystem of claim 1, wherein the non-volatile storage comprises non-volatile solid state memory devices.

12. The storage subsystem of claim 1, wherein the reduction in buffer size reduces the risk of data loss by reducing an amount of time elapsed before data in the write buffer is written to the non-volatile storage.

13. A method performed by a storage subsystem to handle write operations, the method comprising:

receiving write data from a host system, and storing the write data in a variable-size write buffer implemented in a volatile memory of the storage subsystem;

moving the write data from the write buffer to non-volatile storage of the storage subsystem, and signaling to the host system when the write buffer is available to receive additional write data; and

automatically adjusting a size of the write buffer over time based at least in part on one or more monitored conditions of the storage subsystem that are reflective of a likelihood of a data loss in the storage subsystem, so as to vary a sustained write speed of the storage subsystem and an associated risk of data loss, the automatically adjusting comprising reducing the size of the write buffer between at least a first non-zero size value and a second non-zero size value in response to the one or more monitored conditions indicating an increased risk of data loss.

14. The method of claim 13, wherein the one or more monitored conditions comprise a temperature reading.

15. The method of claim 13, wherein the one or more monitored conditions comprise a consistency of a power signal received from the host system.

16. The method of claim 13, wherein the one or more monitored conditions comprise an elapsed time since the storage subsystem was last powered up.

17. The method of claim 13, wherein the one or more monitored conditions comprise an output of a vibration sensor.

18. The method of claim 13, wherein the one or more monitored conditions comprise data regarding a current wear level of the non-volatile storage.

19. The method of claim 13, wherein the storing the write data in the write buffer and the moving the write data from the write buffer to non-volatile storage occur simultaneously.

20. The method of claim 13, wherein the reduction in buffer size reduces the risk of data loss by reducing an amount of time elapsed before write data in the write buffer is written to the non-volatile storage.