Data storage device adjusting a power loss threshold based on samples of supply voltage

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims priority to provisional U.S. Patent Application Ser. No. 61/968,841, filed on Mar. 21, 2014, which is hereby incorporated by reference in its entirety.

BACKGROUND

Data storage devices, such as disk drives and solid state drives are employed in numerous areas such as computer systems (e.g., desktops, laptops, portables, etc.) and consumer devices (e.g., music players, cell phones, cameras, etc.). User data is typically stored in a non-volatile memory, such as a magnetic disk or a non-volatile semiconductor memory (e.g., Flash memory). When a power loss event occurs, the data storage device may execute an emergency operation such as flushing write data stored in a write cache to the non-volatile memory to avoid data loss, or parking a head on a ramp to avoid damaging the head.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1A shows a data storage device comprising a non-volatile memory powered by a supply voltage.

FIG. 1B is a flow diagram according to an embodiment wherein the supply voltage is sampled, and a power loss threshold is adjusted based on the supply voltage samples.

FIG. 2 shows control circuitry according to an embodiment wherein an analog comparator compares the supply voltage to the power loss threshold.

FIG. 3 shows control circuitry according to an embodiment wherein a power loss event is detected when the supply voltage falls below the power loss threshold for a predetermined interval.

FIG. 4 shows control circuitry according to an embodiment wherein a digital comparator compares the supply voltage samples to the power loss threshold.

FIG. 5 shows control circuitry according to an embodiment wherein a power loss event is detected when a predetermined number of consecutive supply voltage samples fall below the power loss threshold.

FIG. 6 shows an embodiment wherein the power loss threshold is adjusted based on an average and standard deviation of the supply voltage samples.

FIG. 7A shows a data storage device in the form of a disk drive according to an embodiment.

FIG. 7B shows a data storage device in the form of a solid state drive according to an embodiment.

FIG. 7C shows a data storage device in the form of a hybrid drive according to an embodiment.

FIG. 8 is a flow diagram according to an embodiment wherein the control circuitry of a disk drive positions a head near an outer diameter of a disk based on the power loss threshold.

FIG. 9 shows an embodiment wherein a size of a write cache is adjusted based on the power supply threshold.

FIG. 10 shows an embodiment wherein a percentage of write cache data that is flushed to a non-volatile semiconductor memory is based on the power loss threshold.

DETAILED DESCRIPTION

FIG. 1A shows a data storage device 2 according to an embodiment comprising a non-volatile memory 4 and control circuitry 6 configured to execute the flow diagram of FIG. 1B. A supply voltage 8 for powering the non-volatile memory is received (block 10) and sampled to generate supply voltage samples (block 12). A power loss threshold is adjusted based on the supply voltage samples (block 14), and a power loss event is detected based on the power loss threshold and the supply voltage (block 16), wherein when the power loss event is detected (block 18) an emergency operation is executed (block 20).

In one embodiment, when a power loss event is detected the data storage device may continue to function normally until the supply voltage falls below a “drop dead” threshold. Accordingly, there is an operating margin of supply voltage as measured from the power loss threshold to the drop dead threshold as illustrated in FIG. 6. During the interval where the supply voltage is falling through this operating margin toward the drop dead threshold, the data storage device may perform an emergency operation such as flushing a write cache before parking a head in a disk drive or flushing a write cache in a solid state drive. Accordingly, in one embodiment adjusting the power loss threshold based on the samples of the supply voltage increases the average operating margin of the supply voltage, thereby improving the performance of the data storage device on average while ensuring a safe power-down operation during periods when the operating margin is decreased.

The supply voltage 8 may be processed relative to the power loss threshold in any suitable manner. FIG. 2 shows control circuitry according to an embodiment wherein a suitable sampling device 22 (e.g., an analog-to-digital converter) samples the supply voltage 8 to generate supply voltage samples 24. Block 26 processes the supply voltage samples 24 to adjust a power loss threshold 28 which is compared to the supply voltage 8 at analog comparator 30. In this embodiment, a power loss event 32 is detected when the supply voltage 8 falls below the power loss threshold 28. In another embodiment, the supply voltage 8 may be filtered using a suitable analog filter (e.g., a low-pass filter) in order to attenuate high frequency noise, wherein the filtered supply voltage is compared to the power loss threshold 28 at comparator 30.

FIG. 3 shows control circuitry according to another embodiment wherein the output of the analog comparator 30 enables a timer 34 when the supply voltage 8 falls below the power loss threshold 28. A power loss event 32 is detected when the output 36 of the timer 34 exceeds a threshold 38 at comparator 40. In this embodiment, the supply voltage 8 must remain below the power loss threshold 38 for an interval determined by the threshold 38 before a power loss event 32 is detected. This embodiment has a similar effect as filtering the supply voltage 8 with a low pass filter, and in one embodiment both a low pass filter and a timer may be employed to reduce the number of false power loss events detected.

FIG. 4 shows control circuitry according to an embodiment wherein a power loss event 32 is detected when one of the supply voltage samples 24 falls below the power loss threshold 28 at digital comparator 42. In one embodiment, the supply voltage samples 24 may be filtered using any suitable digital filter (e.g., a low-pass filter) and the filtered supply voltage samples compared to the power loss threshold 28 at the digital comparator 42.

FIG. 5 shows control circuitry according to an embodiment wherein the output of the digital comparator 42 enables a timer 44 when the supply voltage samples 24 fall below the power loss threshold 28. A power loss event 32 is detected when the output 46 of the timer 44 exceeds a threshold 48 at comparator 50. In this embodiment, at least two consecutive supply voltage samples 24 must remain below the power loss threshold 38 (as determined by threshold 48) before a power loss event 32 is detected. This embodiment has a similar effect as filtering the supply voltage samples 24 with a low pass filter, and in one embodiment both a low pass filter and a timer may be employed to reduce the number of false power loss events detected.

The power loss threshold may be adjusted in any suitable manner based on the supply voltage samples. FIG. 6 shows an embodiment wherein the power loss threshold is adjusted by computing the average u_k of the supply voltage samples and then subtracting N times the standard deviation σ of the supply voltage samples. In one embodiment, the average u_k and the standard deviation σ may be computed over a moving window of the supply voltage samples. As illustrated in the example of FIG. 6, the power loss threshold is decreased as the noise in the supply voltage increases, and then the power loss threshold is increased as the noise in the supply voltage decreases. This embodiment reduces the likelihood that noise and/or transients in the supply voltage samples will trigger a false power loss event while ensuring true power loss events are still detected. FIG. 6 illustrates how adjusting the power loss threshold based on the level of noise/transients in the supply voltage samples increases the average operating margin of the supply voltage, thereby increasing the average interval available during a power loss event to execute the emergency operation. During periods of high supply voltage noise/transients, the operating margin decreases with a corresponding decrease in the interval available to execute the emergency operation. As described in greater detail below, in one embodiment the data storage device may be configured into different operating modes based on the level of the power loss threshold and the corresponding interval available to execute the emergency operation during a power loss event.

In the embodiment of FIG. 6, the control circuitry 6 adjusts the power loss threshold in quantized steps, such as in steps of at least ten millivolts, in order to maintain a sufficient margin between the power loss threshold and the supply voltage, thereby avoiding a decrease in performance due to detecting false power loss events. The power loss threshold may be decreased down to the minimum threshold shown in FIG. 6 which corresponds to a minimum operating margin for the supply voltage and a corresponding minimum interval to execute the emergency operation before the supply voltage drops to the drop dead threshold.

The embodiments of the present invention may be employed in any suitable data storage device 2 comprising any suitable non-volatile memory. FIG. 7A shows a data storage device comprising a disk drive including a head 52 actuated over a disk 54 and control circuitry 56. FIG. 7B shows a data storage device comprising a solid state drive including a plurality of non-volatile semiconductor memories 58A, 58B, etc., such as flash memories, and control circuitry 60. FIG. 7C shows a data storage device comprising a hybrid drive comprising components of a disk drive shown in FIG. 7A combined with the non-volatile semiconductor memory 62 such as shown in FIG. 7B. In an embodiment described below, the non-volatile semiconductor memory 62 may be used to flush at least part of a write cache when a power loss event is detected.

In one embodiment the data storage device may be configured into different operating modes based on the level of the power loss threshold and the corresponding interval available to execute the emergency operation during a power loss event. For example, in one embodiment the data storage device may comprise a disk drive such as shown in FIG. 7A, wherein the control circuitry 56 may configure the disk drive into different operating modes relative to the time needed to park the head 52 onto a ramp 64 during a power loss event. An example of this embodiment is illustrated in the flow diagram of FIG. 8 wherein when the power loss threshold is reduced based on the supply voltage samples (block 66), the head 52 is positioned near an outer diameter of the disk 54. When configured into this mode, the control circuitry 56 may deny or constrain access to the disk 54, such as by denying or constraining host commands, and/or by servicing host commands (when possible) by reading from a non-volatile semiconductor memory 62 such as shown in FIG. 7C or by caching write data in a write cache located in the non-volatile semiconductor memory 62 or near an outer diameter of the disk. Constraining the head 52 to an outer band of the disk 54 reduces the seek time and power needed to park the head 52 onto the ramp 64 at block 72 when a power loss event is detected at block 70.

In one embodiment the data storage device may comprise a write cache for caching write data received from a host in connection with executing write commands. The write cache may improve performance of the data storage device by temporarily storing write data in a volatile semiconductor memory (e.g., DRAM), thereby enabling the host to transmit a stream of write data uninterrupted by the latency of the non-volatile memory 4. The data storage device may then flush the write data from the write cache to the non-volatile memory as a background task, such as during an idle time. The data storage device may also flush the write cache to the non-volatile memory 4 as part of the emergency operation executed when a power loss event is detected so that the write data is not lost. FIG. 9 illustrates an embodiment wherein the size of the write cache may be adjusted based on the power loss threshold. For example, as the power loss threshold is decreased thereby decreasing the interval available during a power loss event, the size of the write cache may be decreased. The smaller write cache helps ensure the entire write cache may be flushed to the non-volatile memory 4 during a power loss event. As the power loss threshold is increased, the size of the write cache may be increased in order increase the performance of the data storage device. Although FIG. 9 shows a linear relationship of write cache size relative to the power loss threshold, the relationship may be represented using any suitable function, such as a suitable polynomial.

FIG. 10 illustrates an embodiment wherein the data storage device comprising a hybrid drive such as shown in FIG. 7C, and a percentage of write cache data that is flushed to a non-volatile semiconductor memory 62 during a power loss event is based on the power loss threshold. For example, as the power loss threshold is decreased the percentage of the write cache data flushed to the non-volatile semiconductor memory 62 is increased since the access latency of the non-volatile semiconductor memory 62 is typically less than the access latency of the disk 54. Although FIG. 10 shows a linear relationship, the relationship may be represented using any suitable function, such as a suitable polynomial. In one embodiment, the control circuitry 56 may compute an execution time needed to flush the write cache data to either the disk 54 or the non-volatile semiconductor memory 62, and then allocate the percentage of write cache data to each non-volatile memory based on the power loss threshold and the corresponding interval available to flush the write cache during a power loss event.

Any suitable control circuitry may be employed in the embodiments of the present invention, such as one or more integrated circuits. In one embodiment, the control circuitry comprises a microprocessor executing instructions, the instructions being operable to cause the microprocessor to perform the flow diagrams described herein. The instructions may be stored in any computer-readable medium. In one embodiment, they may be stored on a non-volatile semiconductor memory external to the microprocessor, or integrated with the microprocessor in a SOC. In another embodiment, the instructions are stored in a non-volatile memory and read into a volatile semiconductor memory when the data storage device is powered on. In yet another embodiment, the control circuitry comprises suitable logic circuitry, such as state machine circuitry.

The various features and processes described above may be used independently of one another, or may be combined in various ways. All possible combinations and subcombinations are intended to fall within the scope of this disclosure. In addition, certain method, event or process blocks may be omitted in some implementations. The methods and processes described herein are also not limited to any particular sequence, and the blocks or states relating thereto can be performed in other sequences that are appropriate. For example, described tasks or events may be performed in an order other than that specifically disclosed, or multiple may be combined in a single block or state. The example tasks or events may be performed in serial, in parallel, or in some other manner. Tasks or events may be added to or removed from the disclosed example embodiments. The example systems and components described herein may be configured differently than described. For example, elements may be added to, removed from, or rearranged compared to the disclosed example embodiments.

While certain example embodiments have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the inventions disclosed herein. Thus, nothing in the foregoing description is intended to imply that any particular feature, characteristic, step, module, or block is necessary or indispensable. Indeed, the novel methods and systems described herein may be embodied in a variety of other forms; furthermore, various omissions, substitutions and changes in the form of the methods and systems described herein may be made without departing from the spirit of the embodiments disclosed herein.