RAID random distribution scheme

A storage system comprising a control interface configured to logically represent a storage space constituted by a plurality of storage devices as at least one RAID group configured for storing a plurality of RAID stripes;

• each RAID stripe comprises a group of data segments arranged in a first subset of data segments and a second subset of data segments;
• the first subset is protected by a first type-1 syndrome segment, the second subset is protected by a second type-1 syndrome segment and the entire group of data segments is protected by a type-2 syndrome segment;
• a control layer comprises a RAID controller configured to:
• generate a randomized distribution scheme for randomly distributing the data segments and syndrome segments in a RAID stripe to respective storage devices in the RAID group;
• and, responsive to a need to write data to the RAID group, to:
• generate at least one RAID stripe;
• and utilize the randomized distribution scheme for distributing the segments in the at least one RAID stripe to the storage devices in the RAID group, such that such any two failed storage devices in the RAID group would have the same first probability (P) for being protected by the same type-1 syndrome segment and the same second probability (1-P) for each one of the two failed storage devices being protected by a different type-1 syndrome segment, irrespective of the relative location of the two failed storage devices;
• wherein the RAID controller is configured for generating the randomized distribution scheme to:
• generate a randomization table, wherein each row in the randomization table comprises a sequence of integers in random order, the sequence of integers in each row representing one possible randomized distribution of segments in a RAID stripe across storage devices in the RAID group;
• select a row from the randomization table;
• and distribute the segments in the RAID stripe to the storage devices in the RAID group based on a respective sequence of integers in the selected row.

A non-transitory program storage device readable by a computer, tangibly embodying a computer readable code portion executable by the computer to perform a method of allocating segments in a RAID stripe to respective storage devices in a corresponding RAID group;

• each RAID stripe comprises a group of data segments arranged in a first subset of data segments and a second subset of data segments;
• the first subset is protected by a first type-1 syndrome segment, the second subset is protected by a second type-1 syndrome segment and the entire group of data segments is protected by a type-2 syndrome segment;
• the method comprising:
• generating a randomized distribution scheme for randomly distributing the data segments and syndrome segments in a RAID stripe to respective storage devices in the RAID group, including:
• generating a randomization table, wherein each row in the randomization table comprises a sequence of integers in random order, the sequence of integers in each row representing one possible randomized distribution of segments in a RAID stripe across storage devices in the RAID group,
• selecting a row from the randomization table, and
• distributing the segments in the RAID stripe to the storage devices in the RAID group based on a respective sequence of integers in the selected row;
• and, responsive to a need to write data to the RAID group:
• generating at least one RAID stripe;
• and using the randomized distribution scheme for distributing the segments in the at least one RAID stripe to the storage devices in the RAID group, such that any two failed storage devices in the RAID group would have the same first probability (P) for being protected by the same type-1 syndrome segment and the same second probability (1-P) for each one of the two failed storage devices being protected by a different type-1 syndrome segment, irrespective of the relative location of the two failed storage devices.