Mass storage device and method of operating the same to store parity data
AssigneeSkyera, LLC
InventorsRadoslav Danilak, Guiqiang Dong, Ladislav Steffko
Life of this patent
Estimating the term…
Abstract
A mass storage memory device is disclosed. The device includes a plurality of blades where two blades are used to store parity data corresponding to data stored in the other blades. The device also includes a controller configured to write data to the blades along stripes extending from the other blades to the two blades, where the parity data within a stripe is based on the data written to the other blades in the stripe, and wherein the parity data includes two or more types of parity data.
Claims20
A mass storage device, comprising:
- a plurality of blades, wherein two of the blades are configured to exclusively store parity data and the other blades are configured to store data written to the device;
- and
- a controller configured to write data to the blades along stripes extending from the other blades to the two blades, and to write parity data to the two blades, wherein the parity data within a stripe is based on the data written to the other blades in the stripe, and wherein the parity data includes two types of parity data.
A method of writing data to a mass storage device comprising a plurality of blades, each blade comprising a plurality of pages, the method comprising:
- receiving data to be stored;
- writing data corresponding to the received data along a stripe of the mass storage device to a first subset of the blades;
- calculating first and second types of parity data, wherein the parity data is calculated based on the written data;
- and
- writing the first and second types of calculated parity data along the stripe to a second subset of the blades, wherein the second subset of blades is configured to exclusively store parity data.
A method of operating a mass storage device comprising a plurality of blades, the method comprising:
- configuring at least two blades to exclusively store parity data;
- configuring the other blades to store data written to the device;
- providing a controller configured to write data to the blades along stripes extending from the other blades to the at least two blades, and to write parity data to the at least two blades, wherein the parity data within a stripe is based on the data written to the other blades in the stripe, and wherein the parity data includes two types of parity data
- receiving an indication that one of the other blades has failed;
- receiving an instruction to access data located on the failed blade;
- recovering data of the failed blade;
- and
- executing the instruction using the recovered data.
A method of operating a mass storage device comprising a plurality of blades and a controller, the method comprising:
- configuring at least two blades to exclusively store parity data;
- the controller operating the device with all of the blades;
- the controller receiving an indication that one of the blades has failed, based on parity data stored in the at least two blades configured to exclusively store parity data;
- the controller operating the device with the non-failing blades;
- the controller receiving an indication that the failed blade has been replaced;
- and
- the controller operating the device with all of the blades.
A method of operating a mass storage device comprising a plurality of blades, the method comprising:
- determining an age of the device;
- and
- selecting an error recovery method based on the age of the device, where in the
- selected error recovery method is selected from the group consisting of ECC, RAID, and bad column mapping.
A method of operating a mass storage device comprising a plurality of blades, the method comprising:
- configuring at least two blades to exclusively store parity data;
- operating the device with all of the blades;
- receiving an indication that one of the blades is at least partially unavailable, based on parity data stored in the at least two blades configured to exclusively store parity data;
- operating the device with the available blades;
- and
- recovering data of the unavailable blade based on data stored in the available blades.
Dependencies are inferred from the claim text.