Method and apparatus for providing distributed checkpointing

A controller comprising logic configured to:

• determine an amount of checkpointing storage memory space needed for checkpointing state data corresponding to an application;
• and
• checkpoint the state data corresponding to the application into the amount of checkpointing storage memory space by distributing the state data among available non-volatile random access memories (NVRAMs) in a plurality of nodes in a network based on the determination.

A controller comprising logic configured to:

• obtain a checkpointing memory allocation request message from a node in a plurality of nodes in a network;
• and
• provide checkpointing by directing the checkpointing memory allocation request message from the node to at least another node in the plurality of nodes in the network to distribute state data corresponding to an application for the node among the plurality of nodes in the network.