Data redundancy in a cluster system

A node apparatus for use in a cluster system which is configured to store data and at least one of data replicas or erasure-coded segments of data across a plurality of nodes in the cluster system, based at least in part on a redundancy policy, the node apparatus comprising:

• one or more processors;
• and
• one or more storage devices coupled to the one or more processors, wherein:
• the node apparatus is configured to connect to one or more networks connected to the plurality of nodes of the cluster system, and to send and receive input/output (I/O) operations associated with its one or more storage devices;
• the node apparatus is configured to act as a first node of the cluster system which determines, upon a change of the redundancy policy, updated configuration information based on the change of the redundancy policy, and which sends a first message indicating the change of the redundancy policy to other nodes of the plurality of nodes;
• and/or
• the node apparatus is configured to act as a second node of the cluster system which, upon receiving the first message from a node acting as the first node, is configured to determine whether data stored on its own one or more storage devices is affected by the change of the redundancy policy, based at least in part on configuration information which is indicative of a data placement of the data and the data replicas or erasure-coded segments on the plurality of nodes.

One or more non-transitory computer readable media having executable instructions stored thereon for execution in a cluster system, which comprises a plurality of nodes connected to one or more networks, and which is configured to store data, data replicas, and/or erasure-coded segments of data across the plurality of nodes in the cluster system, based at least in part on a redundancy policy, each of the plurality of nodes comprising one or more storage devices, and each of the plurality of nodes configured to access its own one or more storage devices and to send and receive input/output (I/O) operations associated with its one or more storage devices, the executable instructions, when executed, causing one or more of the nodes to perform operations comprising:

• providing configuration information which is indicative of a data placement of data and data replicas or erasure-coded segments on the plurality of nodes, to each of the plurality of nodes;
• determining, by a first node of the plurality of nodes, upon a change of the redundancy policy, updated configuration information based on the change of the redundancy policy;
• and
• sending, by the first node, a first message including information indicating the change of the redundancy policy from the first node to other nodes of the plurality of nodes, wherein at least one of the other nodes is configured to determine its own updated configuration information based on the change of the redundancy policy included in the first message.

A method for controlling a cluster system which comprises a plurality of nodes connected to one or more networks and which is configured to store data and at least one of data replicas or erasure-coded segments of data across a plurality of nodes in the cluster system, based at least in part on a redundancy policy, each of the plurality of nodes comprising one or more storage devices, and each of the plurality of nodes configured to access its own one or more storage devices and to send and receive input/output (I/O) operations associated with its one or more storage devices, the method comprising:

• providing configuration information which is indicative of a data placement of the data and data replicas or erasure-coded segments on the plurality of nodes, to each of the plurality of nodes;
• determining, by a node of the cluster system acting as a first node upon a change of the redundancy policy, updated configuration information based on the change of the redundancy policy;
• and
• sending a first message including information indicating the change of the redundancy policy from the first node to other nodes, wherein at least one of the other nodes is configured to determine its own updated configuration information based on the change of the redundancy policy included in the first message.