Storage system

1. A storage system that improves a rebuild of a spared drive and prevents degradation of I/O performance, the storage system comprising:

a plurality of nodes interconnected via a network, wherein each of the plurality of nodes is configured to provide volumes;

a plurality of storage drives that are interconnected via the network; and

a management system that is communicatively coupled to the plurality of nodes and the plurality of storage drives,

wherein each of the plurality of storage drives is accessible from one or more of the plurality of nodes without intervention of the other nodes, and

wherein the management system or one of the plurality of nodes is configured to determine a number of storage drives to be included in a new distribution range based on access frequency to a first volume and based on a rebuild time of each storage drive included in the new distribution range, wherein the new distribution range distributes data of the first volume.

2. The storage system according to claim 1 , wherein the management system or the one of the plurality of nodes is configured to determine that the new distribution range comprises storage drives accessible from an owner node providing the first volume without intervention of the other nodes in a case where the access frequency is equal to or higher than a first threshold.

3. The storage system according to claim 1 , wherein the management system or the one of the plurality of nodes is configured to:

compare the access frequency with a plurality of thresholds to determine a range including the access frequency; and

determine a number of storage drives associated with the determined range to be a number of storage drives to be included in the new distribution range.

4. The storage system according to claim 1 , wherein the management system or the one of the plurality of nodes is configured to preferentially keep storage drives accessible from an owner node providing the first volume without intervention of the other nodes in a case where the determined number of storage drives to be included in the new distribution range is smaller than a number of storage drives included in a current distribution range.

5. The storage system according to claim 1 , wherein the management system or the one of the plurality of nodes is configured to determine storage drives to be removed from a current distribution range of the first volume based on sizes of free spaces of storage drives included in the current distribution range.

6. The storage system according to claim 1 , wherein the management system or the one of the plurality of nodes is configured to:

estimate a rebuild time under a condition of a number of storage drives in a current distribution range and a rebuild time under a condition of a number of storage drives in the new distribution range in a case where the determined number of storage drives to be included in the new distribution range is greater than the number of storage drives included in the current distribution range; and

determine to maintain the current distribution range in a case where the estimated rebuild time under the condition of the number of storage drives in the new distribution range is longer than the estimated rebuild time under the condition of the number of the storage drives in the current distribution range.

7. The storage system according to claim 6 , wherein the estimating the rebuild times refers to throughputs of storage drives in each distribution range and throughputs of interfaces between nodes.

8. The storage system according to claim 1 , wherein the management system or the one of the plurality of nodes is configured to determine storage drives to be added to a current distribution range of the first volume based on sizes of free spaces of storage drives not included in the current distribution range.

9. The storage system according to claim 1 , wherein the management system or the one of the plurality of nodes is configured to:

determine that the new distribution range comprises storage drives accessible from an owner node providing the first volume without intervention of the other nodes in a case where the access frequency is equal to or higher than a first threshold;

estimate a rebuild time under a condition of a number of storage drives in a current distribution range and a rebuild time under a condition of a number of storage drives in the new distribution range in a case where the determined number of storage drives to be included in the new distribution range is greater than the number of storage drives included in the current distribution range; and

determine to maintain the current distribution range in a case where the estimated rebuild time under the condition of the number of storage drives in the new distribution range is longer than the estimated rebuild time under the condition of the number of storage drives in the current distribution range.

10. A method of controlling a storage system including a plurality of storage drives and a plurality of nodes interconnected via a network and configured to provide volumes,

each of the plurality of storage drives being accessible from one or more of the plurality of nodes without intervention of the other nodes, and

the method comprising:

monitoring access frequency to a first volume; and

determining a number of storage drives to be included in a new distribution range to distribute data of the first volume based on a result of monitoring the access frequency and based on a rebuild time of each storage drive included in the new distribution range, wherein the new distribution range distributes data of the first volume.