Storing data in a distributed storage system
View Patent ↗According to examples, a storage node may include storage devices and a controller that may determine whether all of a plurality of data chunks of a first intra-node portion of a stripe have been stored on the storage node. Based on a determination that all of the data chunks have been stored, a first intra-node parity chunk may be stored at a second one of the storage devices, in which the first intra-node parity chunk may be determined from at least one of the data chunks of the first intra-node portion. Based on a determination that at least one of the data chunks has not been stored, storage of a first intra-node parity chunk of the stripe on the storage node may be delayed until a determination is made that all of the data chunks of the first intra-node portion have been stored at the storage node.
1. A method comprising:
storing, by a processor, a first data chunk belonging to a stripe on a first storage device of a plurality of storage devices of a first storage node of a plurality of storage nodes, wherein the stripe is to be stored across the plurality of storage nodes in a plurality of intra-node portions, including a first intra-node portion to be stored on the first storage node and that comprises a plurality of data chunks including the first data chunk;
before all of the data chunks of the first intra-node portion have been stored on the first storage node:
storing a first inter-node parity chunk on a storage device of a second storage node of the plurality of storage nodes, wherein the first inter-node parity chunk is determined from at least the first data chunk; and
delaying storage of a first intra-node parity chunk at the first storage node; determining whether all of the data chunks of the first intra-node portion of the stripe have been stored on the first storage node; and
when all of the data chunks of the first intra-node portion of the stripe have been stored on the first storage node, storing the first intra-node parity chunk at a second one of the storage devices of the first storage node, wherein the first intra-node parity chunk is determined from at least one of the data chunks of the first intra-node portion of the stripe.
2. The method of claim 1 , comprising: storing the stripe across the plurality of storage nodes in a redundant manner using erasure coding.
3. The method of claim 1 , comprising:
storing the plurality of intra-node portions of the stripe on respective storage nodes of the plurality of storage nodes, including storing the first intra-node portion on the first storage node.
4. The method of claim 3 , wherein, for each of one or more of the plurality of intra-node portions, the intra-node portion comprises a plurality of data chunks and at least one intra-node parity chunk, the first intra-node portion comprising the first intra-node parity chunk.
5. The method of claim 4 , wherein at least one of the plurality of intra-node portions comprises at least one inter-node parity chunk and at least one intra-node parity chunk determined from at least the at least one inter-node parity chunk of the intra-node portion, the intra-node portion stored on the second storage node comprising the first inter-node parity chunk.
6. The method of claim 5 , wherein:
each of the plurality of storage nodes comprises a respective plurality of storage devices; and
for each of the plurality of intra-node portions stored on respective storage nodes of the plurality of storage nodes, each chunk of the intra-node portion in stored on a different storage device of the respective storage node.
7. The method of claim 5 , comprising:
storing inter-node parity chunks synchronously relative to the storage of respective data chunks of the stripe; and
storing inter-node parity chunks asynchronously relative to the storage of respective data chunks of the stripe.
8. A non-transitory storage medium comprising instructions executable by a processor of a storage system controller to:
store a first data chunk belonging to a stripe on a first storage device of a plurality of storage devices of a first storage node of a plurality of storage nodes, wherein the stripe is to be stored across the plurality of storage nodes in a plurality of intra-node portions, including a first intra-node portions to be stored on the first storage node and that comprises a plurality of data chunks including the first data chunk;
before a threshold amount of the data chunks of the first intra-node portion have been stored on the first storage node:
store a first inter-node parity chunk on a storage device of a second storage node of the plurality of storage nodes, wherein the first inter-node parity chunk to determined from at least the first data chunk; and
delay storage of a first intra-node parity chunk at the first storage node; determine whether the threshold amount of the data chunks of the first intra-node portion of the stripe have been stored on the first storage node; and
when the threshold amount of the data chunks of the first intra-node portion of the stripe have been stored on the first storage node, store the first intra-node parity chunk at a second one of the storage devices of the first storage node, wherein the first intra-node parity chunk is determined from at least one of the data chunks of the first intra-node portion of the stripe.
9. The storage medium of claim 8 , wherein the threshold amount of the data chunks of the first intra-node portion of the stripe is all of the data chunks of the first intra-node portion.
10. The storage medium of claim 8 , wherein the threshold amount of the data chunks of the first intra-node portion of the stripe is less than all of the data chunks of the first intra-node portion.
11. The storage medium of claim 8 , wherein the threshold amount of the data chunks of the first intra-node portion of the stripe is one of 50 percent, 75 percent, and 90 percent of the data chunks of the first intra-node portion.
12. The storage medium of claim 8 , the instructions comprising instructions to:
store the plurality of intra-node portions of the stripe on respective storage nodes of the plurality of storage nodes, including storing the first intra-node portion on the first storage node.
13. The storage medium of claim 12 , wherein, for each of one or more of the plurality of intra-node portions, the intra-node portion comprises a plurality of data chunks and at least one intra-node parity chunk, the first intra-node portion comprising the first intra-node parity chunk.
14. The storage medium of claim 13 , wherein at least one of the plurality of intra-node portions comprises at least one inter-node parity chunk and at least one intra-node parity chunk determined from at least the at least one inter-node parity chunk of the intra-node portion, the intra-node portion stored on the second storage node comprising the first inter-node parity chunk.
15. The storage medium of claim 14 , the instructions comprising instructions to:
store inter-node parity chunks synchronously relative to the storage of respective data chunks of the stripe; and
store inter-node parity chunks asynchronously relative to the storage of respective data chunks of the stripe.
16. A storage system controller comprising:
a processor; and
a non-transitory storage medium comprising instructions executable by the processor to:
store a first data chunk belonging to a stripe on a first storage device of a plurality of storage devices of a first storage node of a plurality of storage nodes, wherein the stripe is to be stored across the plurality of storage nodes in a plurality of intra-node portions, including a first intra-node portion to be stored on the first storage node and that comprises a plurality of data chunks including the first data chunk;
before a threshold amount of the data chunks of the first intra-node portion have been stored on the first storage node:
store a first inter-node parity chunk on a storage device of a second storage node of the plurality of storage nodes, wherein the first inter-node parity chunk is determined from at least the first data chunk; and
delay storage of a first intra-node parity chunk at the first storage node; determine whether the threshold amount of the data chunks of the first intra-node portion of the stripe have been stored on the first storage node; and
when the threshold amount of the data chunks of the first intra-node portion of the stripe have been stored on the first storage node, store the first intra-node parity chunk at a second one of the storage devices of the first storage node, wherein the first intra-node parity chunk is determined from at least one of the data chunks of the first intra-node portion of the stripe.
17. The storage system controller of claim 16 , wherein the threshold amount of the data chunks of the first intra-node portion of the stripe is all of the data chunks of the first intra-node portion.
18. The storage system controller of claim 16 , wherein the threshold amount of the data chunks of the first intra-node portion of the stripe is less than all of the data chunks of the first intra-node portion.
19. The storage system controller of claim 16 , the instructions comprising instructions to:
store the plurality of intra-node portions of the stripe on respective storage nodes of the plurality of storage nodes, including storing the first intra-node portion on the first storage node.
20. The storage system controller of claim 19 , wherein, for each of one or more of the plurality of intra-node portions, the intra-node portion comprises a plurality of data chunks and at least one intra-node parity chunk, the first intra-node portion comprising the first intra-node parity chunk.