摘要:
Data in less than all of the pages of a non-volatile memory block are updated by programming the new data in unused pages of either the same or another block. In order to prevent having to copy unchanged pages of data into the new block, or to program flags into superceded pages of data, the pages of new data are identified by the same logical address as the pages of data which they superceded and a time stamp is added to note when each page was written. When reading the data, the most recent pages of data are used and the older superceded pages of data are ignored. This technique is also applied to metablocks that include one block from each of several different units of a memory array, by directing all page updates to a single unused block in one of the units.
摘要:
Techniques for producing and supplying various voltage levels within a memory system having multiple memory blocks (e.g., memory chips) are described. The various voltage levels can be produced by voltage generation circuitry (e.g., charge pump and/or regulator circuitry) within the memory system. The various voltage levels can be supplied to the multiple memory blocks through a power bus. According to one aspect, charge pump and/or regulator circuits are provided within at most one of the memory blocks of a memory system (unless back-ups are provided for fault tolerance), and a power bus is used to distribute the generated voltage levels to other of the memory blocks. According to another aspect, a memory controller generates multiple supply voltage levels that are distributed (e.g., via a power bus) to each of the memory blocks.
摘要:
A non-volatile memory system is formed of floating gate memory cells arranged in blocks as the smallest unit of memory cells that are erasable together. The system includes a number of features that may be implemented individually or in various cooperative combinations. One feature is the storage in separate blocks of the characteristics of a large number of blocks of cells in which user data is stored. These characteristics for user data blocks being accessed may, during operation of the memory system by its controller, be stored in a random access memory for ease of access and updating. According to another feature, multiple sectors of user data are stored at one time by alternately streaming chunks of data from the sectors to multiple memory blocks. Bytes of data in the stream may be shifted to avoid defective locations in the memory such as bad columns. Error correction codes may also be generated from the streaming data with a single generation circuit for the multiple sectors of data. The stream of data may further be transformed in order to tend to even out the wear among the blocks of memory. Yet another feature, for memory systems having multiple memory integrated circuit chips, provides a single system record that includes the capacity of each of the chips and assigned contiguous logical address ranges of user data blocks within the chips which the memory controller accesses when addressing a block, making it easier to manufacture a memory system with memory chips having different capacities. A typical form of the memory system is as a card that is removably connectable with a host system but may alternatively be implemented in a memory embedded in a host system. The memory cells may be operated with multiple states in order to store more than one bit of data per cell.
摘要:
A buffer cache interposed between a non-volatile memory and a host may be partitioned into segments that may operate with different policies. Cache policies include write-through, write and read-look-ahead. Write-through and write back policies may improve speed. Read-look-ahead cache allows more efficient use of the bus between the buffer cache and non-volatile memory. A session command allows data to be maintained in volatile memory by guaranteeing against power loss.
摘要:
A non-volatile memory is divided into logical zones by the card controller in order reduce the size of the data structures it uses for address translation. Zone boundaries are adjusted to accommodate defects allowed by memory test to improve card yields and to adjust boundaries in the field to extend the usable lifetime of the card. Firmware scans for the presence of defective blocks on the card. Once the locations of these blocks are known, the firmware calculates the zone boundaries in such a way that good blocks are equally distributed among the zones. Since the number of good blocks meets the card test criteria by the memory test criteria, defects will reduce card yield fallout. The controller can perform dynamic boundary adjustments. When defects occur, the controller can perform the analysis again and, if needed, redistributes the zone boundaries, moving any user data.
摘要:
A non-volatile memory system is formed of floating gate memory cells arranged in blocks as the smallest unit of memory cells that are erasable together. One feature is the storage in separate blocks of the characteristics of a large number of blocks of cells in which user data is stored. These characteristics for user data blocks being accessed may, during operation of the memory system by its controller, be stored in a random access memory for ease of access and updating. A typical form of the memory system is as a card that is removably connectable with a host system but may alternatively be implemented in a memory embedded in a host system. The memory cells may be operated with multiple states in order to store more than one bit of data per cell.
摘要:
Data in less than all of the pages of a non-volatile memory block are updated by programming the new data in unused pages of either the same or another block. In order to prevent having to copy unchanged pages of data into the new block, or to program flags into superceded pages of data, the pages of new data are identified by the same logical address as the pages of data which they superceded and a time stamp is added to note when each page was written. When reading the data, the most recent pages of data are used and the older superceded pages of data are ignored. This technique is also applied to metablocks that include one block from each of several different units of a memory array, by directing all page updates to a single unused block in one of the units.
摘要:
A portable memory storage device is disclosed where access to information on the device is granted only upon proper biometric authentication of a user. The device includes a controller, a non-volatile memory which may be a flash memory, and a biometric scanner system for controlling access to the information within the non-volatile memory. Each of the controller, non-volatile memory and biometric scanner system may be mounted in a base of the portable device, with the biometric system having an exposed surface on a top portion of the base for accepting biometric data such as a fingerprint. A cover is provided which includes a USB connector capable of mating within a USB port of the host device to establish communications between the portable and host devices. The cover also covers the exposed portion of the biometric scanner to protect the sensor when the portable memory storage device is not in use.
摘要:
In order to maintain the integrity of data stored in a flash memory that are susceptible to being disturbed by operations in adjacent regions of the memory, disturb events cause the data to be read, corrected and re-written before becoming so corrupted that valid data cannot be recovered. The sometimes conflicting needs to maintain data integrity and system performance are balanced by deferring execution of some of the corrective action when the memory system has other high priority operations to perform. In a memory system utilizing very large units of erase, the corrective process is executed in a manner that is consistent with efficiently rewriting an amount of data much less than the capacity of a unit of erase.
摘要:
Data are encoded using convolutional coding prior to storage in a nonvolatile memory array, so that errors that occur when the data are read may be corrected even where there is a large number of such errors. Coding rates of less than one increase the amount of data to be stored but allow correction of large numbers of errors.