Data storage method and apparatus

1. A data storage method in a distributed storage system comprising a computing node and a plurality of storage nodes, the method performed by the computing node, the method comprising:

receiving a data storage request comprises to-be-stored data;

determining a hash partition to store the to-be-stored data, wherein the hash partition comprises a first storage node group and a second storage node group, the first storage node group comprises R storage nodes that belong to a first data center, the second storage node group comprises R storage nodes that belong to a second data center, and R is an integer greater than 2;

dividing the to-be-stored data into N data slices, and obtaining M check slices of the N data slices, wherein N and M are positive integers, and N+M=R; and

separately writing the N data slices and the M check slices into the R storage nodes in the first storage node group and the R storage nodes in the second storage node group.

2. The method according to claim 1 , wherein the data storage request further comprises a key, the method further comprising:

determining based on the key, the hash partition to store the to-be-stored data.

3. The method according to claim 1 , wherein the hash partition is determined from the first data center and the second data center by a management node in the distributed storage system based on a fault tolerance level of the distributed storage system.

4. The method according to claim 1 , further comprising:

writing the N data slices and the M check slices into the R storage nodes in the first storage node group; and

copying, into the R storage nodes in the second storage node group, the N data slices and the M check slices that are stored in the R storage nodes in the first storage node group.

5. The method according to claim 4 , further comprising:

repairing a slice that fails to be written into a storage node in the first storage node group when a quantity of slices that fail to be written into the first storage node group is less than or equal to M; and

copying, into the R storage nodes in the second storage node group, the repaired slice.

6. The method according to claim 4 , further comprising:

repairing a slice that fails to be copied into the second storage node group when a quantity of slices that fail to be copied into the second storage node group is less than or equal to M.

7. A computing node in a distributed storage system comprising the computing node and a plurality of storage nodes, the computing node comprising:

a communications interface configured to receive a data storage request comprising to-be-stored data; and

a processor coupled to the communications interface and configured to:

determine a hash partition to store the to-be-stored data,

divide the to-be-stored data into N data slices, and obtain M check slices of the N data slices, and

separately write the N data slices and the M check slices into R storage nodes in a first storage node group and R storage nodes in a second storage node group, wherein the hash partition comprises the first storage node group and the second storage node group, the first storage node group comprises R storage nodes that belong to a first data center, the second storage node group comprises R storage nodes that belong to a second data center, R is an integer greater than 2, N and M are positive integers, and N+M=R.

8. The computing node according to claim 7 , wherein the data storage request further comprises a key, and the processor is further configured to determine, based on the key, the hash partition to store the to-be-stored data.

9. The computing node according to claim 7 , wherein the processor is further configured to determine the hash partition from the first data center and the second data center based on a fault tolerance level of the distributed storage system.

10. The computing node according to claim 7 , wherein the processor is further configured to:

write the N data slices and the M check slices into the R storage nodes in the first storage node group; and copy, into the R storage nodes in the second storage node group, the N data slices and the M check slices that are stored in the R storage nodes in the first storage node group.

11. The computing node according to claim 10 , wherein the processor is further configured to:

repair a slice that fails to be written into a storage node in the first storage node group when a quantity of slices that fail to be written into the first storage node group is less than or equal to M; and

copy, into the R storage nodes in the second storage node group, the repaired slice.

12. The computing node according to claim 10 , wherein the processor is further configured to:

repair a slice that fails to be written into the second storage node group when a quantity of slices that fail to be copied into the second storage node group is less than or equal to M.