Storage unit generation of redundancy information in a redundant storage array system

Claim 2

A redundant storage array system comprising an array controller for receiving a write command, and a plurality of failure independent data storage units for storing data and redundancy information in the form of blocks, wherein at least one of the storage units includes a local processing means for locally generating and locally storing a new redundancy block derived from a corresponding new data block, a corresponding old data block, and a corresponding old redundancy block by combining the old data block, the new data block, and the old redundancy block within the local processing means.

Claim 3

The redundant storage array system of claim 1, wherein each local processing means further includes means for: a. receiving an old data block and a new data block in a single transmission; b. reading an old redundancy block previously stored in the storage unit associated with the processing means, the old redundancy block corresponding to the old data block; and c. storing the new redundancy block in the storage unit associated with the processing means.

Claim 4

The redundant storage array system of claim 2, wherein: a. the old redundancy block is an old parity block; b. the new redundancy block is a new parity block; and c. the processing means further includes means for generating the new parity block by an exclusive-OR operation performed on the old data block, the new data block, and the old parity block.

Claim 5

The redundant storage array system of claim 1, wherein the redundancy information is generated using a Reed-Solomon code.

Claim 6

The redundant storage array system of claim 1, further comprising an array controller coupled to the plurality of failure independent data storage units, wherein the array controller includes a control means for: a. receiving the new data block from a central processing unit; b. receiving the old data block from a first data storage unit; c. transmitting the new data block to the first data storage unit; and d. transmitting the new data block and the old data block in a single transmission to a second data storage unit, the second data storage unit having a local processing means.

Claim 7

The redundant storage array system of claim 5, wherein: a. the old redundancy block is an old parity block; b. the new redundancy block is a new parity block; and c. the processing means further includes means for generating the new parity block by an exclusive-OR operation performed on the old data block, the new data block, and the old parity block.

Claim 8

The redundant storage array system of claim 2, further including an array controller coupled to the plurality of failure independent data storage units, wherein the array controller includes a control means for: a. receiving the new data block; b. receiving the old data block from a first data storage unit; c. transmitting the new data block to the first data storage unit; and d. transmitting the new data block and the old data block in a single transmission to a second data storage unit, the second data storage unit having a local processing means.

Claim 9

The redundant storage array system of claim 7, wherein: a. the old redundancy block is an old parity block; b. the new redundancy block is a new parity block; and c. the processing means further includes means for generating the new parity block by an exclusive-OR operation performed on the old data block, the new data block, and the old parity block.

Claim 10

The redundant storage array system of claim 1, wherein the processing means further includes means for: a. receiving a sum block generated from the new data block and the old data block; and b. reading an old redundancy block previously stored in the storage unit associated with the processing means, the old redundancy block corresponding to the old data block; and wherein the new redundancy block is locally generated from the sum block and the old redundancy block.

Claim 11

The redundant storage array system of claim 9, wherein: a. the old redundancy block is an old parity block; b. the new redundancy block is a new parity block; and c. the processing means for generating the new parity block further includes means for performing an exclusive-OR operation upon the sum block and the old parity block.

Claim 12

The redundant storage array system of claim 9, further including an array controller coupled to a central processing unit and the plurality of failure independent data storage units, wherein the array controller includes a control means for: a. receiving the new data block from a central processing unit; b. receiving the old data block from a first data storage unit; c. transmitting the new data block to the first data storage unit; d. generating the sum block from the old data block and the new data block; and e. transmitting the sum block to a second data storage unit having a local processing means.

Claim 13

The redundant storage array system of claim 11, wherein: a. the old redundancy block is an old parity block; b. the new redundancy block is a new parity block; c. the control means further includes means for generating the sum block by an exclusive-OR operation performed on the old data block and the new data block; and d. the processing means further includes means for generating the new parity block by an exclusive-OR operation performed on the sum block and the old parity block.

Claim 14

The redundant storage array system of claim 1, further including an array controller coupled to the plurality of failure independent data storage units, wherein the array controller includes a control means for: a. receiving the new data block from a central processing unit; b. receiving the old data block from a first data storage unit; c. generating a sum block from the new data block and the old data block; d. transmitting the sum block to a second data storage unit having a local processor; and e. transmitting the new data block to the first data storage unit; wherein the new redundancy block is locally generated from the sum block and the old redundancy block.

Claim 15

The redundant storage array system of claim 13, wherein: a. the old redundancy block is an old parity block; b. the new redundancy block is a new parity block; c. the control means further includes means for generating the sum block by an exclusive-OR operation performed on the old data block and the new data block; and d. the processing means further includes means for generating the new parity block by an exclusive-OR operation performed on the sum block and the old parity block.

Claim 16

In a redundant storage array system including an array controller for receiving a write command, and a plurality of failure independent data storage units in which at least one data storage unit includes a local processing means for creating redundancy blocks, a method for improving the generation of redundancy information and the storage of data and redundancy information, including the step of locally generating and locally storing a new redundancy block derived from a corresponding new data block, a corresponding old data block, and a corresponding old redundancy block by combining the old data block, the new data block, and the old redundancy block within the local processing means.

Claim 17

In a redundant storage array system including an array controller for receiving a write command, and a plurality of failure independent data storage units in which at least one data storage unit includes a local processing means for creating redundancy blocks, a method for improving the generation of redundancy information and the storage of data and redundancy information, including the step of locally generating and locally storing a new redundancy block derived from a corresponding new data block, a corresponding old data block, and a corresponding old redundancy block by combining the old data block, the new data block, and the old redundancy block within the local processing means.

Claim 18

The method of claim 15, wherein the generation and storing of a new redundancy block includes the steps of: a. receiving, in a single transmission, the old data block and the new data block into a first data storage unit, the first data storage unit having a local processing means; b. reading the old redundancy block corresponding to the old data block from the first data storage unit; c. creating the new redundancy block from the old data block, the new data block, and the old redundancy block; d. storing the new redundancy block in the first data storage unit; and e. storing the new data block in a second data storage unit.

Claim 19

The method of claim 16, wherein: a. the old redundancy block is an old parity block; b. the new redundancy block is a new parity block; c. an exclusive-OR operation is used to generate the new parity block from the old data block, the new data block, and the old parity block.

Claim 20

The method of claim 15, wherein the redundancy information is generated using a Reed-Solomon code.

Claim 21

In a redundant storage array system including an array controller for receiving a write command, and a plurality of failure independent data storage units for storing data and redundancy information in the form of blocks, at least one of the storage units including a local processing means for creating error correction code blocks, a method for improving the generation of redundancy information and the storage of data and redundancy information, including the steps of: a. reading an old data block from a first data storage unit; b. receiving a sum block which is a composite of a new data block and the old data block in a second data storage unit, the second data storage unit having a processing means for creating error correction code blocks; c. reading an old error correction code block corresponding to the old data block from the second data storage unit; d. creating a new error correction code block within the second data storage unit from the old error correction code block and the sum block; e. storing the new error correction code block in the second data storage unit; and f. storing the new data block in the first data storage unit.

Claim 22

In a redundant storage array system including an array controller for receiving a write command, and a plurality of failure independent data storage units for storing data and redundancy information in the form of blocks, at least one of the storage units including a local processing means for creating error correction code blocks, a method for improving the generation of redundancy information and the storage of data and redundancy information, including the steps of: a. reading an old data block from a first data storage unit; b. receiving a sum block which is a composite of a new data block and the old data block in a second data storage unit, the second data storage unit having a processing means for creating error correction code blocks; c. reading an old error correction code block corresponding to the old data block from the second data storage unit; d. creating a new error correction code block within the second data storage unit from the old error correction code block and the sum block; e. storing the new error correction code block in the second data storage unit; and f. storing the new data block in the first data storage unit.

Claim 23

The method of claim 19, further including the step of generating a sum block within an array controller by an exclusive-OR operation performed upon the old data block and the new data block.

Claim 24

The method of claim 19, wherein: a. the old error correction code block is an old parity block; b. the new error correction code block is a new parity block; c. the sum block is generated by an exclusive-OR operation performed upon the old data block and the new data block; and d. the new parity block is generated by an exclusive-OR operation performed upon the old parity block and the sum block.

Claim 25

The method of claim 19, wherein the redundancy information is generated using a Reed-Solomon code.

Claim 26

A redundant storage array system comprising an array controller for receiving a write command, and a plurality of failure independent data storage units for storing data and redundancy information in the form of blocks, wherein: a. at least one of the storage units is a disk-type storage unit including: (1) a plurality of read/write means, at least one read/write means for reading data from the storage unit and at least one read/write means for writing data to the storage unit; and (2) a local processing means for locally generating and locally storing a new redundancy block derived from a corresponding new data block, a corresponding old data block, and a corresponding old redundancy block by combining the old data block, the new data block, and the old redundancy block within the local processing means.

Claim 27

A redundant storage array system comprising an array controller for receiving a write command, and a plurality of failure independent data storage units for storing data and redundancy information in the form of blocks, wherein: a. at least one of the storage units is a disk-type storage unit including: (1) a plurality of read/write means, at least one read/write means for reading data from the storage unit and at least one read/write means for writing data to the storage unit; and (2) a local processing means for locally generating and locally storing a new redundancy block derived from a corresponding new data block, a corresponding old data block, and a corresponding old redundancy block by combining the old data block, the new data block, and the old redundancy block within the local processing means.

Claim 28

The redundant storage array system of claim 23, wherein each local processing means further includes means for: a. receiving an old data block and a new data block in a single transmission; b. reading an old redundancy block previously stored in the storage unit associated with the processing means with the first read/write means, the old redundancy block corresponding to the old data block; and c. writing the new redundancy block in the storage unit associated with the processing means with the second read/write means.

Claim 29

The redundant storage array system of claim 24, wherein: a. the old redundancy block is an old parity block; b. the new redundancy block is a new parity block; and c. the processing means further includes means for generating the new parity block by an exclusive-OR operation performed on the old data block, the new data block, and the old parity block.

Claim 30

The redundant storage array system of claim 23, wherein the redundancy information is generated using a Reed-Solomon code.

Claim 31

The redundant storage array system of claim 23, further comprising an array controller coupled to the plurality of failure independent data storage units, wherein the array controller includes a control means for: a. receiving the new data block from a central processing unit; b. receiving the old data block from a first data storage unit; c. transmitting the new data block to the first data storage unit; and d. transmitting the new data block and the old data block in a single transmission to a second data storage unit, the second data storage unit having a local processing means.

Claim 32

The redundant storage array system of claim 27, wherein: a. the old redundancy block is an old parity block; b. the new redundancy block is a new parity block; and c. the processing means further includes means for generating the new parity block by an exclusive-OR operation performed on the old data block, the new data block, and the old parity block.

Claim 33

The redundant storage array system of claim 24, further including an array controller coupled to the plurality of failure independent data storage units, wherein the array controller includes a control means for: a. receiving the new data block; b. receiving the old data block from a first data storage unit; c. transmitting the new data block to the first data storage unit; and d. transmitting the new data block and the old data block in a single transmission to a second data storage unit, the second data storage unit having a local processing means.

Claim 34

The redundant storage array system of claim 29, wherein: a. the old redundancy block is an old parity block; b. the new redundancy block is a new parity block; and c. the processing means further includes means for generating the new parity block by an exclusive-OR operation performed on the old data block, the new data block, and the old parity block.

Claim 35

The redundant storage array system of claim 23, wherein the processing means further includes means for: a. receiving a sum block generated from the new data block and the old data block; and b. reading an old redundancy block previously stored in the storage unit associated with the processing means with the first read/write means, the old redundancy block corresponding to the old data block; and wherein the new redundancy block is locally generated from the sum block and the old redundancy block and written with the second read/write means to the storage means associated with the processing means.

Claim 36

The redundant storage array system of claim 31, wherein: a. the old redundancy block is an old parity block; b. the new redundancy block is a new parity block; and c. the processing means for generating the new parity block further includes means for performing an exclusive-OR operation upon the sum block and the old parity block.

Claim 37

The redundant storage array system of claim 31, further including an array controller coupled to a central processing unit and the plurality of failure independent data storage units, wherein the array controller includes a control means for: a. receiving the new data block from a central processing unit; b. receiving the old data block from a first data storage unit; c. transmitting the new data block to the first data storage unit; d. generating the sum block from the old data block and the new data block; and e. transmitting the sum block to a second data storage unit having a local processing means.

Claim 38

The redundant storage array system of claim 33, wherein: a. the old redundancy block is an old parity block; b. the new redundancy block is a new parity block; c. the control means further includes means for generating the sum block by an exclusive-OR operation performed on the old data block and the new data block; and d. the processing means further includes means for generating the new parity block by an exclusive-OR operation performed on the sum block and the old parity block.

Claim 39

The redundant storage array system of claim 23, further including an array controller coupled to the plurality of failure independent data storage units, wherein the array controller includes a control means for: a. receiving the new data block from a central processing unit; b. receiving the old data block from a first data storage unit; c. generating a sum block from the new data block and the old data block; d. transmitting the sum block to a second data storage unit having a local processor; and e. transmitting the new data block to the first data storage unit; wherein the new redundancy block is locally generated from the sum block and the old redundancy block.

Claim 40

The redundant storage array system of claim 35, wherein: a. the old redundancy block is an old parity block; b. the new redundancy block is a new parity block; c. the control means further includes means for generating the sum block by an exclusive-OR operation performed on the old data block and the new data block; and d. the processing means further includes means for generating the new parity block by an exclusive-OR operation performed on the sum block and the old parity block.

Claim 41

In a redundant storage array system including a plurality of failure independent data storage units in which at least one data storage unit includes a local processing means for creating redundancy blocks, at least a first read/write means for reading data from the storage unit, and at least a second read/write means for writing data to the storage unit, a method for improving the generation of redundancy information and the storage of data and redundancy information, including the step of: a. locally generating a new redundancy block from a corresponding new data block, a corresponding old data block, and a corresponding old redundancy block read by the first read/write means; and b. writing the new redundancy block to the data storage unit with the second read/write means.

Claim 42

In a redundant storage array system including a plurality of failure independent data storage units in which at least one data storage unit includes a local processing means for creating redundancy blocks, at least a first read/write means for reading data from the storage unit, and at least a second read/write means for writing data to the storage unit, a method for improving the generation of redundancy information and the storage of data and redundancy information, including the step of: a. locally generating a new redundancy block from a corresponding new data block, a corresponding old data block, and a corresponding old redundancy block read by the first read/write means; and b. writing the new redundancy block to the data storage unit with the second read/write means.

Claim 43

The method of claim 37, wherein the generation and storing of a new redundancy block includes the steps of: a. receiving, in a single transmission, the old data block and the new data block into a first data storage unit, the first data storage unit having a local processing means, a first read/write means and a second read/write means; b. reading with the first read/write means the old redundancy block corresponding to the old data block from the first data storage unit; c. creating the new redundancy block from the old data block, the new data block, and the old redundancy block; d. writing with the second read/write means the new redundancy block in the first data storage unit; and e. storing the new data block in a second data storage unit.

Claim 44

The method of claim 38, wherein: a. the old redundancy block is an old parity block; b. the new redundancy block is a new parity block; c. an exclusive-OR operation is used to generate the new parity block from the old data block, the new data block, and the old parity block.

Claim 45

The method of claim 37, wherein the redundancy information is generated using a Reed-Solomon code.

Claim 46

In a redundant storage array system including a plurality of failure independent data storage units for storing data and redundancy information in the form of blocks, at least one of the storage units including a local processing means for creating error correction code blocks, a first read/write means for reading data, and a second read/write means for writing data, a method for improving the generation of redundancy information and the storage of data and redundancy information, including the steps of: a. reading an old data block from a first data storage unit; b. receiving a sum block which is a composite of a new data block and the old data block in a second data storage unit, the second data storage unit having a processing means for creating error correction code blocks, a first read/write means for reading data, and a second read/write means for writing data; c. reading with the first read/write means an old error correction code block corresponding to the old data block from the second data storage unit; d. creating a new error correction code block within the second data storage unit from the old error correction code block and the sum block; e. writing with the second read/write means the new error correction code block in the second data storage unit; and f. storing the new data block in the first data storage unit.

Claim 47

In a redundant storage array system including a plurality of failure independent data storage units for storing data and redundancy information in the form of blocks, at least one of the storage units including a local processing means for creating error correction code blocks, a first read/write means for reading data, and a second read/write means for writing data, a method for improving the generation of redundancy information and the storage of data and redundancy information, including the steps of: a. reading an old data block from a first data storage unit; b. receiving a sum block which is a composite of a new data block and the old data block in a second data storage unit, the second data storage unit having a processing means for creating error correction code blocks, a first read/write means for reading data, and a second read/write means for writing data; c. reading with the first read/write means an old error correction code block corresponding to the old data block from the second data storage unit; d. creating a new error correction code block within the second data storage unit from the old error correction code block and the sum block; e. writing with the second read/write means the new error correction code block in the second data storage unit; and f. storing the new data block in the first data storage unit.

Claim 48

The method of claim 41, further including the step of generating a sum block within an array controller by an exclusive-OR operation performed upon the old data block and the new data block.

Claim 49

The method of claim 41, wherein: a. the old error correction code block is an old parity block; b. the new error correction code block is a new parity block; c. the sum block is generated by an exclusive-OR operation performed upon the old data block and the new data block; and d. the new parity block is generated by an exclusive-OR operation performed upon the old parity block and the sum block.

Claim 50

The method of claim 41, wherein the redundancy information is generated using a Reed-Solomon code.