Method and system of multi-fold data protection for high-density storage appliance with improved availability and robustness
AssigneeAlibaba Group Holding Limited
InventorShu Li
Life of this patent
Estimating the term…
Abstract
A first set of data is encoded using a first code to obtain a first-code codeword which includes the first set of data and first-code parity information. The first set of data is stored on a plurality of drives, wherein the first set of data is distributed amongst the plurality of drives. A second set of data is encoded using a second code to obtain a second-code codeword which includes the second set of data and second-code parity information. The second-code codeword is stored on the plurality of drives, wherein the second set of data and second-code parity information are distributed amongst the plurality of drives.
Claims21
A system for storing data, comprising:
- a processor;
- and
- a memory coupled with the processor, wherein the memory is configured to provide the processor with instructions which when executed cause the processor to:
- encode, using a first code, a first set of data to obtain a first-code codeword which includes the first set of data and first-code parity information;
- store the first set of data on a plurality of drives, wherein the first set of data is distributed amongst the plurality of drives;
- encode, using a second code, a second set of data to obtain a second-code codeword which includes the second set of data and second-code parity information;
- and
- store the second-code codeword on the plurality of drives, wherein the second set of data and second-code parity information are distributed amongst the plurality of drives.
A system for accessing data, comprising:
- a processor;
- and
- a memory coupled with the processor, wherein the memory is configured to provide the processor with instructions which when executed cause the processor to:
- read a first set of data from a plurality of drives, wherein the first set of data is distributed amongst the plurality of drives;
- decode, using a first code, a first-code codeword which includes the first set of data and first-code parity information to obtain a first corrected set of data;
- read a second-code codeword from the plurality of drives, wherein:
- the second-code codeword includes a second set of data and second-code parity information;
- and
- the second set of data and the second-code parity information are distributed amongst the plurality of drives;
- and
- decode, using a second code, the second-code codeword which includes the second set of data and the second-code parity information to obtain a second corrected set of data.
A method for storing data, comprising:
- encoding, using a first code, a first set of data to obtain a first-code codeword which includes the first set of data and first-code parity information;
- storing the first set of data on a plurality of drives, wherein the first set of data is distributed amongst the plurality of drives;
- encoding, using a second code, a second set of data to obtain a second-code codeword which includes the second set of data and second-code parity information;
- and
- storing the second-code codeword on the plurality of drives, wherein the second set of data and second-code parity information are distributed amongst the plurality of drives.
A method for accessing data, comprising:
- reading a first set of data from a plurality of drives, wherein the first set of data is distributed amongst the plurality of drives;
- decoding, using a first code, a first-code codeword which includes the first set of data and first-code parity information to obtain a first corrected set of data;
- reading a second-code codeword from the plurality of drives, wherein:
- the second-code codeword includes a second set of data and second-code parity information;
- and
- the second set of data and the second-code parity information are distributed amongst the plurality of drives;
- and
- decoding, using a second code, the second-code codeword which includes the second set of data and the second-code parity information to obtain a second corrected set of data.
Dependencies are inferred from the claim text.