Managing solid state storage devices
AssigneeEverpure, Inc.
InventorsLINLIN ZHOU, MILLEN KATYAL, ZOLTAN DeWITT, ETHAN MILLER, JOHN COLGROVE
Life of this patent
Estimating the term…
Abstract
An indication that a die of the solid-state storage device has been marked as likely to fail based on a number of failed input/output (I/O) operations performed on the die satisfying a threshold may be received by a storage system controller from a solid-state storage device. In response to receiving the indication, one or more remedial actions associated with the die of the solid-state storage device may be performed.
Claims20
A solid-state storage device comprising:
- a plurality of dies each including a plurality of memory blocks;
- and
- a controller coupled to the plurality of dies, the controller configured to:
- monitor input/output (I/O) operations associated with each of the plurality of dies;
- determine that a first die of the plurality of dies is likely to fail based on a failure metric derived from the monitored I/O operations;
- in response to determining that the first die is likely to fail:
- mark the first die as likely to fail;
- initiate a data recovery operation to recover data stored on the first die prior to failure;
- and
- restrict at least one type of access to the first die while permitting continued operation of one or more other dies of the plurality of dies.
A method comprising:
- monitoring, by a controller of a solid-state storage device, input/output (I/O) operations associated with a plurality of dies;
- determining that a first die of the plurality of dies is likely to fail based on a failure metric associated with the monitored I/O operations;
- marking the first die as likely to fail;
- initiating a recovery operation to recover data stored on the first die;
- and restricting at least one type of access to the first die while continuing to perform storage operations using one or more other dies of the plurality of dies.
A non-transitory computer-readable medium storing instructions that, when executed by a processing device of a solid-state storage device, cause the processing device to:
- monitor input/output (I/O) operations associated with a plurality of dies;
- determine that a first die is likely to fail based on a failure metric;
- mark the first die as likely to fail;
- initiate recovery of data stored on the first die;
- and restrict at least one type of access to the first die while permitting continued use of one or more other dies.
Dependencies are inferred from the claim text.