Claims Analysis
Total Claims: 16
Claim 1
Independent Claim
133 words
A fault tolerant memory system comprising, in combination: a multidimensional array of memory devices arranged in addressable locations, each addressable memory location containing a plurality of data bits and a plurality of check bits for checking the integrity of all the data bits and check bits at a given addressable location; a pool of spare memory devices including at least two spare memory devices which may be assigned to replace any of said memory devices or a previously assigned spare memory device; and means for detecting, in response to said data bits and said check bits read from a given memory location, a failing memory device or previously assigned spare memory device and for assigning a previously unassigned spare memory device to replace the failing memory device or previously assigned spare memory device.
Claim 2
Dependent Claim
133 words
A fault tolerant memory system comprising, in combination: a multidimensional array of memory devices arranged in addressable locations, each addressable memory location containing a plurality of data bits and a plurality of check bits for checking the integrity of all the data bits and check bits at a given addressable location; a pool of spare memory devices including at least two spare memory devices which may be assigned to replace any of said memory devices or a previously assigned spare memory device; and means for detecting, in response to said data bits and said check bits read from a given memory location, a failing memory device or previously assigned spare memory device and for assigning a previously unassigned spare memory device to replace the failing memory device or previously assigned spare memory device.
Claim 3
Dependent Claim
41 words
The fault tolerant memory of claim 1 wherein the number of check bits is determined by the smallest digit r which satisfies the equation k+r<=2.sup.(r-1) where k is the number of data bits and r is the number of check bits.
Claim 4
Dependent Claim
35 words
The fault tolerant memory of claim 1 additionally including error correction means responsive to said data bits and said check bits to correct the data read from the memory when a correctable error is detected.
Claim 5
Dependent Claim
56 words
The fault tolerant memory of claim 1 additionally including means for reading a spare memory unit when it has been assigned to replace an addressed memory element previously determined to have failed and to merge the data read from said assigned spare memory unit into the desired bit position of the data read from the memory.
Claim 6
Dependent Claim
144 words
A fault tolerant memory system comprising, in combination: a multidimensional array of memory devices arranged in addressable locations, each addressable memory location containing a plurality of data bits and a plurality of check bits for checking the integrity of all the data bits and check bits at a given addressable location; a pool of spare memory devices including at least two spare memory devices which may be assigned to replace any identifiable portion of one said memory device or a previously assigned spare memory device; and means for detecting, in response to said data bits and said check bits read from a given memory location, a failing portion of a memory device or a previously assigned spare memory device and for assigning a previously unassigned spare memory device to replace the failing portion of a memory device or a previously assigned spare memory device.
Claim 7
Dependent Claim
144 words
A fault tolerant memory system comprising, in combination: a multidimensional array of memory devices arranged in addressable locations, each addressable memory location containing a plurality of data bits and a plurality of check bits for checking the integrity of all the data bits and check bits at a given addressable location; a pool of spare memory devices including at least two spare memory devices which may be assigned to replace any identifiable portion of one said memory device or a previously assigned spare memory device; and means for detecting, in response to said data bits and said check bits read from a given memory location, a failing portion of a memory device or a previously assigned spare memory device and for assigning a previously unassigned spare memory device to replace the failing portion of a memory device or a previously assigned spare memory device.
Claim 8
Dependent Claim
41 words
The fault tolerant memory of claim 5 wherein the number of check bits is determined by the smallest digit r which satisfies the equation k+r<=2.sup.(r-1) where k is the number of data bits and r is the number of check bits.
Claim 9
Dependent Claim
35 words
The fault tolerant memory of claim 5 additionally including error correction means responsive to said data bits and said check bits to correct the data read from the memory when a correctable error is detected.
Claim 10
Dependent Claim
56 words
The fault tolerant memory of claim 5 additionally including means for reading a spare memory unit when it has been assigned to replace an addressed memory element previously determined to have failed and to merge the data read from said assigned spare memory unit into the desired bit position of the data read from the memory.
Claim 11
Dependent Claim
272 words
A fault tolerant memory system comprising, in combination: a two dimensional array of memory devices arranged in addressable locations, each addressable memory location containing a plurality of data bits and a plurality of check bits for checking the integrity of all the data bits and check bits at a given addressable location; a plurality of memory devices in a pool of spare memory devices; accessing means for reading an addressable location and all said memory devices in said pool; spare merging network responsive to said accessing means to substitute the data read from a previously assigned spare memory device for a previously determined failing bit position in said addressable location to produce memory read data; error checking and correcting facility responsive to said memory read data for detecting errors in said memory read data and for producing corrected memory read data which comprises said memory read data when no error is detected and comprises and correctly altered data when an error is detected, said error checking and correcting facility producing syndrome information When an error is detected to identify the location of the detected error; maintenance facility responsive to said syndrome information to count the errors that occur for each memory element and producing a spare assignment signal when the error count for a given memory device exceeds a predetermined maximum; and spare allocation means responsive to said spare assignment signal to assign a spare memory device to replace the memory device indicated by said spare assignment signal to have failed so that further memory reads or writes will address the assigned spare memory device in place of the failed memory device.
Claim 12
Dependent Claim
272 words
A fault tolerant memory system comprising, in combination: a two dimensional array of memory devices arranged in addressable locations, each addressable memory location containing a plurality of data bits and a plurality of check bits for checking the integrity of all the data bits and check bits at a given addressable location; a plurality of memory devices in a pool of spare memory devices; accessing means for reading an addressable location and all said memory devices in said pool; spare merging network responsive to said accessing means to substitute the data read from a previously assigned spare memory device for a previously determined failing bit position in said addressable location to produce memory read data; error checking and correcting facility responsive to said memory read data for detecting errors in said memory read data and for producing corrected memory read data which comprises said memory read data when no error is detected and comprises and correctly altered data when an error is detected, said error checking and correcting facility producing syndrome information When an error is detected to identify the location of the detected error; maintenance facility responsive to said syndrome information to count the errors that occur for each memory element and producing a spare assignment signal when the error count for a given memory device exceeds a predetermined maximum; and spare allocation means responsive to said spare assignment signal to assign a spare memory device to replace the memory device indicated by said spare assignment signal to have failed so that further memory reads or writes will address the assigned spare memory device in place of the failed memory device.
Claim 13
Dependent Claim
39 words
The fault tolerant memory system of claim 9 wherein said spare allocation means is additionally operative to assign a spare memory unit to replace a previously assigned spare memory unit indicated by said spare assignment signal to have failed.
Claim 14
Dependent Claim
41 words
The fault tolerant memory of claim 9 wherein the number of check bits is determined by the smallest digit r which satisfies the equation k+r<=2.sup.(r-1) where k is the number of data bits and r is the number of check bits.
Claim 15
Independent Claim
191 words
A method for operating a fault tolerant memory system comprising the steps of: reading data bits and check bits from an addressed location in a memory and a corresponding location in each of a pool of spare memory devices, said memory including a plurality of identifiable memory devices for which any one of said pool of spare memory devices can be substituted; merging previously assigned bits read from the pool of spare into the data and check bits previously determined to have failed, to produce memory output data; checking said memory output data to determine if a memory read failure has occurred and correcting said memory output data as required to produce corrected memory output data, said checking step including producing syndrome data indicating the location of the detected failure; and counting, in response to said syndrome data, the number of failures associated with each memory device of the memory for which one of said spares can be substituted and, when the number of errors exceeds a predefined maximum, indicating that a specific previously unassigned spare should be assigned to replace the identified failing memory device on all subsequent memory accesses.
Claim 16
Independent Claim
191 words
A method for operating a fault tolerant memory system comprising the steps of: reading data bits and check bits from an addressed location in a memory and a corresponding location in each of a pool of spare memory devices, said memory including a plurality of identifiable memory devices for which any one of said pool of spare memory devices can be substituted; merging previously assigned bits read from the pool of spare into the data and check bits previously determined to have failed, to produce memory output data; checking said memory output data to determine if a memory read failure has occurred and correcting said memory output data as required to produce corrected memory output data, said checking step including producing syndrome data indicating the location of the detected failure; and counting, in response to said syndrome data, the number of failures associated with each memory device of the memory for which one of said spares can be substituted and, when the number of errors exceeds a predefined maximum, indicating that a specific previously unassigned spare should be assigned to replace the identified failing memory device on all subsequent memory accesses.