Peer-to-peer redundant array of independent disks (raid) lacking a raid controller

A method comprising:

• taking over for a failed programmable disk drive configured within a peer-to-peer redundant array of independent disks (RAID) including a plurality of other programmable disk drives, by a spare programmable disk drive, the peer-to-peer RAID lacking a RAID controller;
• for a data block of the failed programmable disk drive to be reconstructed by the spare programmable disk drive, sending a request to a first programmable disk drive of the other programmable disk drives, by the spare programmable disk drive;
• in response to the first programmable disk drive receiving the request, sending data relevant to reconstructing the data block from the first programmable disk drive to a next programmable disk drive of the other programmable disk drives;
• performing, by each other programmable disk drive as a given other programmable disk drive, from the next programmable disk drive to a next-to-last programmable disk drive:
• receiving the data relevant to reconstructing the data block from an immediately prior programmable disk drive;
• performing an exclusive-or (XOR) logic operation in relation to the received data relevant to reconstructing the data block and additional data relevant to reconstructing the data block stored by the given other programmable disk drive, yielding XOR'ed data;
• sending the XOR'ed data, as the data relevant to reconstructing the data block, to the other programmable disk drive that is immediately subsequent to the given other programmable disk drive;
• receiving, by the last programmable disk drive of the other programmable disk drives, the data relevant to reconstructing the data block from the next-to-lost programmable disk drive;
• performing, by the last programmable disk drive, the XOR logic operation in relation to the received data relevant to reconstructing the data block and additional data relevant to reconstructing the data block stored by the last programmable disk drive, yielding the data block as reconstructed;
• sending the data block as reconstructed from the last programmable disk drive to the spare programmable disk drive;
• and
• storing the data block as reconstructed by the spare programmable disk drive.

A system comprising:

• a network;
• a plurality of programmable disk drives to communicate with one another over the network and that are organized as a peer-to-peer redundant array of independent disks (RAID) that lacks a RAID controller;
• a first spare programmable disk drive and a second spare programmable disk drive, each of which is to monitor a particular programmable disk drive of the programmable disk drives, and one of which is to replace the particular programmable disk drive within the RAID and is to reconstruct data on the particular programmable disk drive in a specified manner when the particular programmable disk drive has failed,
• wherein in the specified manner, for a data block of the particular programmable disk drive to be reconstructed, each programmable disk drive other than the particular programmable disk drive receives just one piece of data and sends just one piece of data.