Storage system and storage control method
A storage system and storage control method are provided in which a plurality of volumes to be processed by a storage control unit are distributed and evacuated in a normal storage control unit without recovering redundancy of the storage control unit having decreased redundancy, and the storage control unit itself having the decreased redundancy is deleted after the evacuating is completed, and thus reservation information processing resources for guaranteeing recoverability of the redundancy become unnecessary.
1. A storage system comprising a plurality of storage nodes constituting a cluster,
wherein each of the storage nodes includes a storage apparatus storing data, a cluster control unit controlling the entire storage system, and a storage control unit providing a storage area in units of volume to a host apparatus by using the storage apparatus and storing data in the storage apparatus in response to an input/output (I/O) request from the host apparatus,
the storage control unit is configured to constitute a storage control unit group with a storage control unit of another storage node in the cluster, wherein at least one storage control unit of the storage control unit group processes the I/O request from the host apparatus as a storage control unit in an active mode and another storage control unit of the storage control unit group takes over processes of the storage control unit in the active mode, and
the cluster control unit of at least one of the plurality of storage nodes obtains information about a plurality of volumes handled by a first storage control unit group, determines a second storage control unit group with respect to each volume of the plurality of volumes about which the information is obtained, and migrates the plurality of volumes handled by the first storage control unit group from a storage control unit constituting the first storage control unit group to a storage control unit of the second storage control unit group.
2. The storage system according to claim 1 , wherein
the plurality of volumes handled by the first storage control unit group are migrated by selecting a plurality of the second storage control unit groups.
3. The storage system according to claim 1 , wherein
the plurality of volumes are migrated when a storage node related to one storage control unit constituting the first storage control unit group has a failure or is set to be removed.
4. The storage system according to claim 1 , wherein
the cluster control unit is configured such that a cluster control unit of one storage node among the plurality of storage nodes is a cluster control unit of a master role and a cluster control unit of another storage node among the plurality of storage nodes is a cluster control unit of a worker role, and
the cluster control unit of master role manages an operation status of a storage node, storage node management information for managing various information processing resources possessed by a storage node, a pair relationship of storage control units, a relationship between a storage control unit and a storage node, and storage control unit management information for managing an operation status of a storage control unit.
5. The storage system according to claim 1 , wherein
the first storage control unit group is stopped after the plurality of volumes are migrated.
6. The storage system according to claim 1 , wherein
the cluster control unit is configured such that a cluster control unit of one storage node among the plurality of storage nodes is a cluster control unit of a master role and a cluster control unit of another storage node among the plurality of storage nodes is a cluster control unit of a worker role, and
the cluster control unit of master role manages an operation status of a storage node, storage node management information for managing various information processing resources possessed by a storage node, a group relationship of storage control unit, a relationship between a storage control unit and a storage node, and storage control unit management information for managing an operation status of a storage control unit.
7. The storage system according to claim 1 , wherein
each of the plurality of storage nodes further includes a data redundancy unit for creating redundant data with another storage node and storing the redundant data in the storage apparatus,
the storage control unit in the active mode constituting the storage control unit group manages a group relationship of a storage control unit, a relationship between a storage control unit and a storage node, and volume management information for managing an operation status of a storage control unit,
the data redundancy unit manages a relationship between a logical chunk corresponding to a block of volume provided to the host apparatus and a physical chunk obtained by dividing the storage apparatus into small areas, and logical chunk management information for managing a storage control unit group corresponding to a logical chunk, and
a storage control unit in a standby mode constituting the storage control unit group stores a copy of the volume management information.
8. The storage system according to claim 1 , wherein
the storage control unit group includes the storage control unit in the active mode processing the I/O request from the host apparatus and a storage control unit in a standby mode capable of taking over the processes of the storage control unit in the active mode.
9. The storage system of claim 8 , wherein
when an operation mode of a storage control unit existing in a storage node having a failure is the active mode, a remaining storage control unit constituting the storage control unit group switches an operation mode from a standby mode to the active mode and takes over processes of the storage control unit in the active mode before the failure by using volume management information and logical chunk management information.
10. The storage system according to claim 9 , wherein
when the failure occurs, a cluster control unit of a remaining other storage node obtains information about a plurality of volumes handled by a storage control unit group constituted by the storage control unit having the failure, and determines a plurality of storage control unit groups of an evacuating destination with respect to each volume of the plurality of volumes about which the information is obtained, based on any one of a status of a storage node where a storage control unit constituting a storage control unit group of an evacuating destination candidate is arranged, a free space of the storage apparatus, a communication band utilization rate, and a central processing unit (CPU) utilization rate.
11. The storage system according to claim 10 , wherein
when the failure occurs, the cluster control unit of the remaining other storage node obtains information about the plurality of volumes handled by the storage control unit group constituted by the storage control unit having the failure, and collectively determines the plurality of storage control unit groups of the evacuating destination with respect to a plurality of volumes having dependency among the plurality of volumes about which the information is obtained.
12. The storage system according to claim 11 , wherein
when the failure occurs, the cluster control unit of the remaining other storage node, after evacuating of the plurality of volumes handled by the storage control unit group constituted by the storage control unit having the failure is completed, deletes information of the storage control unit group constituted by the storage control unit having the failure from storage control unit management information, and
the cluster control unit of the remaining other storage node releases a CPU and memory assigned to a storage control unit constituting a deleted group.
13. The storage system according to claim 1 , wherein
a volume handled by at least one storage control unit group among the first and second storage control unit groups including a first storage control unit in a first storage node having a failure or set to be removed is migrated and the first storage control unit is stopped, and at the same time, another storage control unit is created in the storage node where the stopped first storage control unit exists and added to another storage control unit group.
14. The storage system according to claim 1 , wherein
a volume handled by a first plurality of storage control unit groups including a first storage control unit in the first storage node having a failure or set to be removed is migrated and the first plurality of storage control units are stopped, and at the same time, another plurality of storage control units are created in a plurality of storage nodes where the stopped first plurality of storage control units exist such that the created another plurality of storage control units form a storage control unit group.
15. A method of controlling a storage system including a plurality of storage nodes constituting a cluster, wherein
each of the plurality of storage nodes includes a storage apparatus storing data, a cluster control unit controlling the entire storage system, and a storage control unit providing a storage area in units of volume to a host apparatus by using the storage apparatus and storing data in the storage apparatus in response to an input/output (I/O) request from the host apparatus,
the storage control unit constitutes a storage control unit group with a storage control unit of another storage node in the cluster, wherein at least one storage control unit of the storage control unit group processes the I/O request from the host apparatus as a storage control unit in an active mode and another storage control unit of the storage control unit group takes over the storage control unit in the active mode, and
when a failure occurs in one storage node among the plurality of storage nodes, the cluster control unit of at least one of the plurality of storage nodes obtains information about a plurality of volumes handled by a first storage control unit group, determines a second storage control unit group with respect to each volume of the plurality of volumes about which the information is obtained, and migrates the plurality of volumes handled by the first storage control unit group from a storage control unit constituting the first storage control unit group to a storage control unit of the second storage control unit group.