Abstract: Multi-write coding of non-volatile memories including a method that receives write data, and a write address of a memory page. The memory page is in either an erased state or a previously written state. If the memory page is in the erased state: selecting a first codeword from a code such that the first codeword encodes the write data and is consistent with a target set of distributions of electrical charge levels in the memory page; and writing the first codeword to the memory page. If the memory page is in the previously written state: selecting a coset from a linear code such that the coset encodes the write data and includes one or more words that are consistent with previously written content of the memory page; selecting a subsequent codeword from the one or more words in the coset; and writing the subsequent codeword to the memory page.

Abstract: Resistive memories having a not-and (NAND) structure including a resistive memory cell. The resistive memory cell includes a resistive memory element for storing a resistance value and a memory element access device for controlling access to the resistive memory element. The memory element access device is connected in parallel to the resistive memory element.

Abstract: A method for error detection in a memory system. The method includes calculating one or more signatures associated with data that contains an error. It is determined if the error is a potential correctable error. If the error is a potential correctable error, then the calculated signatures are compared to one or more signatures in a trapping set. The trapping set includes signatures associated with uncorrectable errors. An uncorrectable error flag is set in response to determining that at least one of the calculated signatures is equal to a signature in the trapping set.

Abstract: Techniques for reading phase change memory that mitigate resistance drift. One contemplated method includes apply a plurality of electrical input signals to the memory cell. The method includes measuring a plurality of electrical output signals from the memory cell resulting from the plurality of electrical input signals. The method includes calculating an invariant component of the plurality of electrical output signals dependent on the configuration of amorphous material in the memory cell. The method also includes determining a memory state of the memory cell based on the invariant component. In one embodiment of the invention, the method further includes mapping the plurality of electrical output signals to a measurements region of a plurality of measurements regions. The measurements regions correspond to memory states of the memory cell.

Abstract: Techniques for reading phase change memory that mitigate resistance drift. One contemplated method includes apply a plurality of electrical input signals to the memory cell. The method includes measuring a plurality of electrical output signals from the memory cell resulting from the plurality of electrical input signals. The method includes calculating an invariant component of the plurality of electrical output signals dependent on the configuration of amorphous material in the memory cell. The method also includes determining a memory state of the memory cell based on the invariant component. In one embodiment of the invention, the method further includes mapping the plurality of electrical output signals to a measurements region of a plurality of measurements regions. The measurements regions correspond to memory states of the memory cell.

Abstract: Techniques for writing to memory using adaptive write techniques. An adaptive write technique includes receiving at a computer a message including a plurality of symbols. The message is written to a memory. The writing to memory includes performing for each symbol in the message: writing a data value to a memory location in the memory and reading contents of the memory location after the data value has been written. The data value is determined at the computer in response to the symbol and to the contents of any memory locations previously read as part of writing the message to the memory. It is determined at the computer if the contents of the memory locations reflect the message. The writing is restarted at the computer in response to determining that the contents of the memory locations do not reflect the message.

Abstract: Techniques for writing to memory using shared address buses. A memory device that includes a plurality of memory arrays connected to a common address bus, the common address bus used to broadcast memory addresses simultaneously to the plurality of memory arrays. Each memory array includes a plurality of memory locations and circuitry for: receiving the broadcasted memory addresses from the address bus; selecting a memory address in the memory array from a list of most recent memory addresses received from the address bus; and performing a memory access at the selected memory address, such that at a given point in time at least two of the memory arrays perform the memory access at a different broadcasted address when the memory access is a write.

Abstract: An analog memory having adjustable write bins including a system for writing to the memory. The system includes a write apparatus interpreting one or more write control signals, generating a write signal, and applying the write signal at a selected memory location to store a desired content. The selected memory location is subject to data dependent noise and is capable of storing a range of values grouped into “n” bins configured such that the average cost to write to at least “n-1” of the bins is within a threshold of a target cost for the selected analog memory location. The system also includes a read apparatus. The system further includes write control circuitry that includes a write signal selector selecting the one or more write control signals responsive to the desired content, current content of the selected memory location, and a bin associated with the desired content.

Abstract: The present invention describes lossless data compression/decompression methods and systems. A random access memory (RAM) operates as a static dictionary and includes commonly used strings/symbols/phrases/words. An input buffer operates as a dynamic dictionary and includes input strings/phrases/symbols/words. A set-associative cache memory operates as a hash table, and includes pointers pointing to the commonly used strings/symbols/phrases/words in the static dictionary and/or pointing to one or more of the input strings/phrases/symbols/words in the dynamic dictionary. Alternatively, the set-associative cache memory combines the dynamic dictionary, the static dictionary and the hash table. When encountering a symbol/phrase/string/word in the static or dynamic dictionary in an input stream, a compressor logic or module places a pointer pointing to the symbol/phrase/string/word at a current location on the output stream.

Abstract: A two-level error control protocol detects errors on the subline level and corrects errors using the codeword for the entire line. This enables a system to read small pieces of coded data and check for errors before accepting them, and in case errors are detected, the whole codeword is read for error correction.

Abstract: Iterative write pausing techniques to improve read latency of memory systems including memory systems with phase change memory (PCM) devices. A PCM device includes a plurality of memory locations and a mechanism for executing an iterative write to one or more of the memory locations in response to receiving a write command that includes data to be written. The executing includes initiating the iterative write, updating a state of the iterative write, pausing the iterative write including saving the state in response to receiving a pause command, and resuming the iterative write in response to receiving a resume command. The resuming is responsive to the saved state and to the data to be written.

Abstract: A content addressable memory using encoded data words and search words, and techniques for operating such device. In one embodiment, the data word is transformed into a code word guaranteeing a mismatch of at least two code word bits of different binary values during the memory search operation when the data word does not match a search word. In another embodiment, the search word is transformed into a search code such that the Hamming distance between the code word and the search code is greater than a given threshold when there is a mismatch of at least one bit between the data word and the search word.

Abstract: A system to improve miscorrection rates in error control code may include an error control decoder with a safe decoding mode that processes at least two data packets. The system may also include a buffer to receive the processed at least two data packets from the error control decoder. The error control decoder may apply a logic OR operation to the uncorrectable error signal related to the processing of the at least two data packets to produce a global uncorrectable error signal. The system may further include a recipient to receive the at least two data packets and the global uncorrectable error signal.

Abstract: A system to improve error correction may include a fast decoder to process data packets until the fast decoder finds an uncorrectable error in a data packet at which point a request for at least two data packets is generated. The system may also include a slow decoder to possibly correct the uncorrectable error in a data packet based upon the at least two data packets.

Abstract: A system to improve error control coding may include memory chips of at least two different kinds. The system may also include error control encoder circuitry to substantially encode data for storage in any memory rank. The system may further include error control decoder circuitry to substantially decode encoded data received from any memory rank. The error decoder circuitry is comprised of a slow decoder and a fast decoder.

Abstract: A system to improve memory failure management may include memory, and an error control decoder to determine failures in the memory. The system may also include an agent that may monitor failures in the memory. The system may further include a table where the error control decoder may record the failures, and where the agent can read and write to.

Abstract: A system to improve error code decoding with retries may include a processing unit that requests data packets, and a queue to hold the data packets for the processing unit. The system may also include a decoder to determine a processing time for each data packet in the queue based upon any errors in each data packet, and if the processing time for a particular data packet is greater than a threshold, then to renew any requests for the data packets that are in the queue.

Abstract: A system to improve error code decoding using historical information may include storage partitioned into memory ranks, and a table to record symbols having failures for each memory rank. The system may also generate a memory rank score for each memory rank. The system may also include an error control decoder that may use the memory rank score when each memory rank is accessed in order to determine whether an error should be corrected or not.

Abstract: A system to improve memory reliability in computer systems that may include memory chips, and may rely on a error control encoder to send codeword symbols for storage in each of the memory chips. At least two symbols from a codeword are assigned to each memory chip and therefore failure of any of the memory chips could affect two symbols or more. The system may also include a table to record failures and partial failures of the codeword symbols for each of the memory chips so the error control encoder can correct subsequent partial failures based upon the previous partial failures. The error control coder is capable of correcting and/or detecting more errors if only a fraction of a chip is noted in the table as having a failure as opposed to a full chip noted as having a failure.

Abstract: Adaptive endurance coding including a method for storing data that includes receiving write data and a write address. A compression algorithm is applied to the write data to generate compressed data. An endurance code is applied to the compressed data to generate a codeword. The endurance code is selected and applied in response to the amount of space saved by applying the compression to the write data. The codeword is written to the write address.