Storage system and data arrangement method of storage system
An object is to construct a storage system with a configuration with a high degree of freedom while ensuring a certain fault tolerance. The storage system includes a plurality of nodes that process an I/O processing request of data. The node has a storage device and a processor that performs the I/O processing on the storage device. The processor constitutes a data redundancy configuration in which data stored in different storage devices is combined as a data set. A management unit for managing the storage system performs a fault tolerance calculation of calculating fault tolerance information for a failure of each component by using component information that is information of a component including at least the and the storage device, and by using data redundancy configuration information related to the data redundancy configuration, and determines a data arrangement by applying the data redundancy configuration related to the calculated fault tolerance information.
1. A storage system comprising:
a plurality of nodes that process an I/O processing request of data, wherein each of the plurality of nodes includes:
a storage device; and
a processor configured to:
perform the I/O processing on the storage device;
use a data redundancy configuration in which data stored in different storage devices is combined as a data set;
perform, when changing a number of the nodes, a fault tolerance calculation of calculating fault tolerance information for a failure of each component based on the number of nodes after the change by using component information that is information of a component including at least one of a node and the storage device, and by using data redundancy configuration information related to the data redundancy configuration; and
determine a data arrangement by applying the data redundancy configuration related to the calculated fault tolerance information,
wherein the fault tolerance information related to the data redundancy configuration includes a number of components having operational continuity in a case where a failure occurs for each of plurality of component types, and
the data arrangement, to which the data redundancy configuration is applied, is determined such that the number of components having operational continuity satisfies a predetermined reference in a case where the failure occurs for each of the plurality of component types.
2. The storage system according to claim 1 , wherein
the processor, in the fault tolerance calculation, sets an arrangement of each piece of data based on the data redundancy configuration information, and calculates the fault tolerance information based on an operational continuity in a case where a failure occurs in each component.
3. The storage system according to claim 1 , wherein
the component includes a power source or a network, and the fault tolerance information includes an operational continuity in a case where a failure occurs in the power source or the network.
4. The storage system according to claim 1 , wherein
the component includes a site, and the fault tolerance information includes an operational continuity in a case where a failure occurs in the site.
5. The storage system according to claim 1 , wherein
the fault tolerance information includes information of a bottleneck region serving as a bottleneck of the data arrangement.
6. The storage system according to claim 1 , wherein
the processor determines the data redundancy configuration and the data arrangement based on the fault tolerance information.
7. The storage system according to claim 1 , wherein
the fault tolerance information is output, and a selection of the data redundancy configuration and the data arrangement is received from a user.
8. A data arrangement method in a storage system including a plurality of nodes that process an I/O processing request of data, wherein
each of the plurality of nodes includes:
a storage device; and
a processor configured to;
perform the I/O processing on the storage device;
use a data redundancy configuration in which data stored in different storage devices is combined as a data set;
perform, when changing a number of the nodes, a fault tolerance calculation of calculating fault tolerance information for a failure of each component based on the number of the nodes after the change by using component information that is information of a component including at least one of a node and the storage device, and by using data redundancy configuration information related to the data redundancy configuration;
determine a data arrangement by applying the data redundancy configuration related to the calculated fault tolerance information,
wherein the fault tolerance information related to the data redundancy configuration includes a number of components having operational continuity in a case where a failure occurs for each of plurality of component types, and
the data arrangement, to which the data redundancy configuration is applied, is determined such that the number of components having operational continuity satisfies a predetermined reference in a case where the failure occurs for each of the plurality of component types.
9. The storage system according to claim 1 , wherein the changing of the number of the nodes is increasing the number of the nodes.
10. The storage system according to claim 1 , wherein the changing of the number of the nodes is decreasing a number of the nodes, and
whether the number of the nodes can be decreased is calculated by determining whether a number of components capable of continuing operation even when a failure occurs for each of the plurality of component types satisfies a predetermined standard based on the number of the nodes after the decrease.