Disk access event control for mapped nodes supported by a real cluster storage system
Disk access event control for mapped nodes of a cluster storage system supporting a redundant array of independent nodes (mapped RAIN) system is disclosed. A mapped RAIN cluster can be allocated on top of one or more real data clusters. In an embodiment, disk access events can be routed via a storage service instance supporting a mapped node. In another embodiment, disk access events can be routed via another storage service instance that does not support the mapped node. Routing the disk access event via another storage service instance that does not support the mapped node can reduce the use of computing resources. Further, the routing of the disk access event can be according to a proportional disk operation value determined based on historical disk access event characteristics.
1. A system, comprising:
a processor; and
a memory that stores executable instructions that, when executed by the processor, facilitate performance of operations, comprising:
determining, by a first storage service instance of a first real node of a real cluster storage system, that a real disk corresponds to a mapped disk that is supported by a second storage service of a second real node of the real cluster storage system, wherein the first real node comprises a first plurality of real disks comprising the real disk, wherein the second real node comprises a second plurality of real disks different than the first plurality of real disks, wherein the second real node comprises a second mapping service instance that is a different mapping service instance than the first mapping service instance, and wherein the first storage service applies a protective mapping scheme, for a mapped node comprising the mapped disk, where the protective mapping scheme prevents a mapping of a primary instance of data and redundant information, which corresponds to the primary instance of the data, to a same real node and thereby reduces a likelihood of a data loss event occurring through inaccessibility to both the primary instance of the data and the redundant information as a result of the same real node becoming inaccessible at a future time;
receiving an indication of a disk access event operation at the first real node; and
routing the disk access event operation to the real disk without traversing the second real node, wherein the routing is supported by the first mapping service instance rather than the second mapping service instance.
2. The system of claim 1 , wherein the determining is based on mapping data of mapped nodes supported by the real cluster system, wherein the mapping of the data is in accord with the protective mapping scheme, and wherein the mapping data is accessed via the first mapping service instance, and wherein the mapped nodes comprise the mapped disk.
3. The system of claim 1 , wherein the real disk is comprised in the first real node.
4. The system of claim 1 , wherein the real disk is comprised in a third real node of the real cluster storage system.
5. The system of claim 1 , wherein the routing is in response to detecting a disk access event corresponding to the disk access event operation.
6. The system of claim 1 , wherein the routing is based on the determining, via the first mapping service instance, that the real disk corresponds to the mapped disk.
7. The system of claim 1 , wherein the routing avoids the disk access event operation traversing the second real node.
8. The system of claim 1 , wherein the operations further comprise determining a proportional disk operation value, and wherein the disk access event is performed according to the proportional disk operation value.
9. The system of claim 8 , wherein the disk operation value is based on an input-output value of historical disk access operations.
10. The system of claim 9 , wherein the input-output value of historical disk access operations reflects an average time to perform a disk access event.
11. The system of claim 10 , wherein the average time to perform a disk access event is a floating window time average value to perform a disk access event.
12. A method, comprising:
determining, by a processor of a first real node of a real cluster storage system executing a first storage service instance and based on mapping data of mapped nodes supported by the real cluster system, that a real disk corresponds to a mapped disk that is supported by a second storage service of a second real node of the real cluster storage system, wherein the real node comprises a first plurality of real disks and a first mapping service instance that is a different mapping service instance than a second mapping service instance comprised in the second real node that comprises a second plurality of real disks different than the first plurality of real disks, wherein the mapping prevents a primary instance of data and corresponding redundant information being mapped to a same real node in accord with a data loss protection rule, and wherein the first mapping service instance enables access to the mapping data local to the first real node instead of being via the second real node; and
initiating, by the processor, the disk access event operation at the real disk.
13. The method of claim 12 , wherein initiating the disk access event operation comprises initiating the disk access event operation at the first real node because the first real node comprises the real disk.
14. The method of claim 12 , wherein the initiating the disk access event operation comprises initiating the disk access event operation at a third real node because the third real node comprises the real disk, and wherein the third real node is comprised in the real cluster storage system.
15. The method of claim 12 , wherein the initiating the disk access event operation comprises bypassing the second storage service instance of the second real node.
16. The method of claim 12 , wherein the initiating the disk access event operation comprises initiating the disk access event operation according to a proportional disk operation value determined from historical disk input-output event characteristics.
17. A non-transitory machine-readable storage medium, comprising executable instructions that, when executed by a processor, facilitate performance of operations, comprising:
in response to detecting a disk access event at a first storage service instance of a first real node of a real cluster storage system, determining that a real disk corresponds to a mapped disk that is supported by a second storage service of a second real node of the real cluster storage system, wherein the first real node comprises a first plurality of real disks that are different real disks than a second plurality of real disks comprised in the second real node, wherein the determining is based on mapping data of mapped nodes supported by the real cluster system that prevents a primary instance of data and corresponding redundant information from being mapped to a same real node in accord with a data loss protection rule, and wherein the mapping data is accessed via a first mapping service instance that is a different mapping service instance than a second mapping service instance executing in the second real node, and wherein the first mapping service instance enables access to the mapping data local to the real node rather than via the second real node; and
indicating that the disk access event operation is to be performed via the real disk without engaging the second storage service instance of the second real node.
18. The non-transitory machine-readable storage medium of claim 17 , wherein the real disk is comprised in the first real node.
19. The non-transitory machine-readable storage medium of claim 17 , wherein the real disk is comprised in a third real node of the real cluster storage system.
20. The non-transitory machine-readable storage medium of claim 18 , wherein the operations further comprise:
determining a proportional disk operation value based on a time windowed average duration of historical disk input-output operations, and wherein the disk access event is performed according to the proportional disk operation value.