Spanned RAID with nested parity
In accordance with embodiments of the present disclosure, a method may be implemented to respond to receiving data A to be stored in a storage system that includes a plurality of N spans, each including a plurality of M drives. The method may include performing data storing operations. The data storing operations may include storing the data A in N portions across the plurality of spans, generating a Za parity based on an XOR operation applied to the N portions of the data A, storing the Za parity in N−1 portions across a subset of the plurality of spans, determining a Zap parity based on an XOR operation applied to the N−1 portions of the Za parity, and storing the Zap parity.
1. A method comprising:
in response to receiving data A to be stored in a storage system that includes a plurality of N spans each including a plurality of M drives, performing data storing operations comprising:
storing the data A in N portions across the plurality of spans;
generating a Za parity based on an XOR operation applied to the N portions of the data A;
storing the Za parity in N−1 portions across a subset of the plurality of spans;
determining a Zap parity that is equal to a result of an XOR operation applied to the N−1 portions of the Za parity; and
storing the Zap parity.
2. The method of claim 1 , wherein the Zap parity is stored on a span that is not within the subset.
3. The method of claim 1 , wherein N is greater than or equal to M+1.
4. The method of claim 1 , wherein a portion of the data A that is stored in a particular span is divided into M−2 portions.
5. The method of claim 4 , further comprising calculating a P parity for the particular span based on an XOR operation applied to the M−2 portions of the data A.
6. The method of claim 1 , wherein the storage system further includes at least one hot spare drive.
7. The method of claim 6 , further comprising: in response to a failure of a particular drive, re-generating data from the particular drive for storage on the at least one hot spare drive.
8. The method of claim 6 , wherein the at least one hot spare drive comprises a plurality of hot spare drives, and wherein the method further comprises: in response to a failure of a particular span, re-generating data from the particular span for storage on the plurality of hot spare drives.
9. An article of manufacture comprising a non-transitory, computer-readable medium having instructions stored thereon that are configured to be executed by a computer system having a storage system that includes a plurality of N spans each including a plurality of M drives, the instructions for:
in response to receiving data A to be stored in the storage system, performing data storing operations comprising:
storing the data A in N portions across the plurality of spans;
generating a Za parity based on an XOR operation applied to the N portions of the data A;
storing the Za parity in N−1 portions across a subset of the plurality of spans;
determining a Zap parity that is equal to a result of an XOR operation applied to the N−1 portions of the Za parity; and
storing the Zap parity.
10. The article of claim 9 , wherein the operations further comprise storing the Zap parity on a span that is not within the subset.
11. The article of claim 9 , wherein N is greater than or equal to M+1.
12. The article of claim 9 , wherein a portion of the data A that is stored in a particular span is divided into M−2 portions.
13. The article of claim 12 , the operations further comprising calculating a P parity for the particular span based on an XOR operation applied to the M−2 portions of the data A.
14. The article of claim 9 , wherein the storage system further includes at least one hot spare drive.
15. A system comprising:
at least one processor; and
a storage system that includes a plurality of N spans each including a plurality of M drives;
wherein, in response to receiving data A to be stored in the storage system, the system is configured to perform data storing operations comprising:
storing the data A in N portions across the plurality of spans;
generating a Za parity based on an XOR operation applied to the N portions of the data A;
storing the Za parity in N−1 portions across a subset of the plurality of spans;
determining a Zap parity that is equal to a result of an XOR operation applied to the N−1 portions of the Za parity; and
storing the Zap parity.
16. The system of claim 15 , wherein the system is further configured to store the Zap parity on a span that is not within the subset.
17. The system of claim 15 , wherein N is greater than or equal to M+1.
18. The system of claim 15 , wherein a portion of the data A that is stored in a particular span is divided into M−2 portions.
19. The system of claim 18 , wherein the system is further configured to calculate a P parity for the particular span based on an XOR operation applied to the M−2 portions of the data A.
20. The system of claim 15 , wherein the storage system further includes at least one hot spare drive.