Multi-storage node system and capacity management method of multi-storage node system

1. A multi-storage node system, comprising:

a plurality of storage nodes,

wherein the plurality of storage nodes each includes a plurality of drives that store data and a control unit that controls writing of data to the plurality of drives,

the control unit divides storage areas of the plurality of drives into a plurality of physical chunks, allocates the divided physical chunks to logical chunks, and stores data,

the plurality of storage nodes are divided into fault sets affected by a single failure, and

the control unit of one primary node that manages the multi-storage node system among the plurality of storage nodes includes a database for managing allocation of the physical chunks of the plurality of storage nodes to the logical chunks for each drive, each storage node, or for each fault set,

wherein the control unit of the primary node

determines whether or not the number of allocated physical chunks of a first drive in a first storage node among the plurality of storage nodes exceeds a sum of the number of unallocated free physical chunks in drives in the first storage node other than the first drive in the first storage node, and

gives an alert for adding another drive, having a number of free physical chunks equal to or larger than the sum of the number of unallocated free physical chunks, to the first storage node when exceeded.

2. The multi-storage node system according to claim 1 , wherein the database includes a drive management table for managing the number of physical chunks, the number of physical chunks allocated to the logical chunks, and the number of physical chunks not allocated to the logical chunks for each drive.

3. The multi-storage node system according to claim 2 , wherein the database includes

an allocated physical chunk (node) management table for managing the number of physical chunks allocated to logical chunks as the number of allocated chunks for each storage node, and

a free physical chunk (node) management table for managing the number of physical chunks not allocated to logical chunks as the number of free physical chunks for each storage node.

4. The multi-storage node system according to claim 3 , wherein the database includes

an allocated physical chunk (fault set) management table for managing the number of physical chunks allocated to the logical chunks as the number of allocated chunks for each fault set, and

a free physical chunk (fault set) management table for managing the number of physical chunks not allocated to logical chunks as the number of free physical chunks for each fault set.

5. The multi-storage node system according to claim 4 , wherein the allocated physical chunk (node) management table and the allocated physical chunk (fault set) management table manage the number of allocated physical chunks for each type of the drive, and

the free physical chunk (node) management table and the free physical chunk (fault set) management table manage the number of physical chunks not allocated to the logical chunks for each type of the drive.

6. The multi-storage node system according to claim 1 , wherein the control unit of the primary node

determines whether or not the number of allocated physical chunks of the first storage node belonging to a first fault set among the plurality of storage nodes exceeds a sum of the number of physical chunks not allocated to the logical chunks in a storage node other than the first storage node belonging to the first fault set, and

gives an alert for adding the number of free physical chunks to the first fault set when exceeded.

7. The multi-storage node system according to claim 6 , wherein the control unit of the primary node

determines whether or not the number of allocated physical chunks of a storage node belonging to a second fault set among the plurality of storage nodes exceeds a sum of the number of physical chunks not allocated to the logical chunks in a storage node belonging to a fault set other than the second fault set, and

gives an alert for adding the number of free physical chunks to the fault set other than the second fault set when exceeded.

8. A capacity management method of a multi-storage node system including a plurality of storage nodes,

the plurality of storage nodes including a plurality of drives that store data and a control unit that controls writing of data to the plurality of drives, the capacity management method comprising:

dividing, by the control unit, storage areas of the plurality of drives into a plurality of physical chunks, allocating the divided physical chunks to logical chunks, and storing data;

dividing the plurality of storage nodes into fault sets affected by a single failure; and

managing, by the control unit of one primary node that manages the multi-storage node system among the plurality of storage nodes, allocation of the physical chunks of the plurality of storage nodes to the logical chunks for each drive, each storage node, or for each fault set by a database,

wherein the control unit of the primary node

determines whether or not the number of allocated physical chunks of a first drive in a first storage node among the plurality of storage nodes exceeds a sum of the number of unallocated free physical chunks in drives in the first storage node other than the first drive in the first storage node, and

gives an alert for adding another drive, having a number of free physical chunks equal to or larger than the sum of the number of unallocated free physical chunks, to the first storage node when exceeded.

9. The capacity management method of the multi-storage node system according to claim 8 , wherein the database includes a drive management table for managing the number of physical chunks, the number of physical chunks allocated to the logical chunks, and the number of physical chunks not allocated to the logical chunks for each drive.

10. The capacity management method of the multi-storage node system according to claim 9 , wherein the database includes

an allocated physical chunk (node) management table for managing the number of physical chunks allocated to logical chunks as the number of allocated chunks for each storage node, and

a free physical chunk (node) management table for managing the number of physical chunks not allocated to logical chunks as the number of free physical chunks for each storage node.

11. The capacity management method of the multi-storage node system according to claim 10 , wherein the database includes

an allocated physical chunk (fault set) management table for managing the number of physical chunks allocated to the logical chunks as the number of allocated chunks for each fault set, and

a free physical chunk (fault set) management table for managing the number of physical chunks not allocated to logical chunks as the number of free physical chunks for each fault set.

12. The capacity management method of the multi-storage node system according to claim 8 , wherein the control unit of the primary node

determines whether or not the number of allocated physical chunks of the first storage node belonging to a first fault set among the plurality of storage nodes exceeds a sum of the number of physical chunks not allocated to the logical chunks in a storage node other than the first storage node belonging to the first fault set, and

gives an alert for adding the number of free physical chunks to the first fault set when exceeded.

13. The capacity management method of the multi-storage node system according to claim 12 , wherein the control unit of the primary node

determines whether or not the number of allocated physical chunks of a storage node belonging to a second fault set among the plurality of storage nodes exceeds a sum of the number of physical chunks not allocated to the logical chunks in a storage node belonging to a fault set other than the second fault set, and

gives an alert for adding the number of free physical chunks to the fault set other than the second fault set when exceeded.