Utilization of erasure codes in a storage system
A plurality of storage nodes in a single chassis is provided. The plurality of storage nodes in the single chassis is configured to communicate together as a storage cluster. Each of the plurality of storage nodes includes nonvolatile solid-state memory for user data storage. The plurality of storage nodes is configured to distribute the user data and metadata associated with the user data throughout the plurality of storage nodes such that the plurality of storage nodes maintain the ability to read the user data, using erasure coding, despite a loss of two of the plurality of storage nodes. A plurality of compute nodes is included in the single chassis, each of the plurality of compute nodes is configured to communicate with the plurality of storage nodes. A method for accessing user data in a plurality of storage nodes having nonvolatile solid-state memory is also provided.
1. A method for accessing user data in a plurality of storage nodes having nonvolatile solid-state memory, comprising:
distributing the user data throughout the plurality of storage nodes using one or more of a set of erasure codes, the one or more erasure codes determined by a plurality of authorities, each authority of the plurality of authorities having ownership for a corresponding range of the user data;
in response to determining that one of the plurality of storage nodes is unreachable;
reading the user data across a remainder of the plurality of storage nodes, using a first subset of erasure codes from the set of erasure codes;
writing the user data across the remainder of the plurality of storage nodes, using a second subset of erasure codes from the set of erasure codes; and
accessing the user data, via one or more of the set of erasure codes as determined by at least one of the plurality of authorities, from the remainder of the plurality of storage nodes.
2. The method of claim 1 , wherein the set of erasure codes includes two or more differing erasure codes coexisting in the plurality of storage nodes.
3. The method of claim 1 , wherein determining that one of the plurality of storage nodes is unreachable is based on one of lack of a heartbeat, lack of a response to a query, or a timeout.
4. The method of claim 1 , wherein the plurality of storage nodes are distributed throughout multiple chassis.
5. The method of claim 1 , wherein the at least one authority is located on a differing storage node from the one of the plurality of storage nodes determined to be unreachable.
6. A tangible, non-transitory, computer-readable media having instructions thereupon which, when executed by a processor, cause the processor to perform a method comprising:
distributing user data throughout a plurality of storage nodes using one or more of a set of erasure codes, the one or more erasure codes determined by a plurality of authorities, each authority of the plurality of authorities having ownership for a corresponding range of the user data;
in response to determining that one of the plurality of storage nodes is unreachable:
reading the user data across a remainder of the plurality of storage nodes, using a first subset of erasure codes from the set of erasure codes;
writing the user data across the remainder of the plurality of storage nodes, using a second subset of erasure codes from the set of erasure codes; and
accessing the user data, via one or more of the set of erasure codes as determined by at least one of the plurality of authorities, from the remainder of the plurality of storage nodes.
7. The computer readable media of claim 6 , wherein the set of erasure codes include two or more differing erasure codes coexisting in the plurality of storage nodes.
8. The computer readable media of claim 6 , wherein determining that one of the plurality of storage nodes is unreachable is based on one of lack of a heartbeat, lack of a response to a query, or a timeout.
9. The computer readable media of claim 6 , wherein the plurality of storage nodes are distributed throughout multiple chassis.
10. The computer readable media of claim 6 , wherein the at least one authority is located on a differing storage node from the one of the plurality of storage nodes determined to be unreachable.
11. A storage cluster, comprising:
solid-state storage memory;
a processor, configurable to execute instructions causing the processor to:
distribute user data throughout a plurality of storage nodes using one or more of a set of erasure codes, the one or more erasure codes determined by a plurality of authorities, each authority of the plurality of authorities having ownership for a corresponding range of the user data;
in response to determining that one of the plurality of storage nodes is unreachable:
reading the user data across a remainder of the plurality of storage nodes, using a first subset of erasure codes from the set of erasure codes;
writing the user data across the remainder of the plurality of storage nodes, using a second subset of erasure codes from the set of erasure codes; and
accessing the user data, via one or more of the set of erasure codes as determined by at least one of the plurality of authorities, from the remainder of the plurality of storage nodes.
12. The storage cluster of claim 11 , wherein the erasure codes include two or more differing erasure codes coexisting in the plurality of storage nodes.
13. The storage cluster of claim 11 , wherein determining that one of the plurality of storage nodes is unreachable is based on one of lack of a heartbeat, lack of a response to a query, or a timeout.
14. The storage cluster of claim 11 , wherein the plurality of storage nodes are distributed throughout multiple chassis.