Showing papers on "Memory refresh published in 2008"

Mini-rank: Adaptive DRAM architecture for improving memory power efficiency

Abstract: The widespread use of multicore processors has dramatically increased the demand on high memory bandwidth and large memory capacity. As DRAM subsystem designs stretch to meet the demand, memory power consumption is now approaching that of processors. However, the conventional DRAM architecture prevents any meaningful power and performance trade-offs for memory-intensive workloads. We propose a novel idea called mini-rank for DDRx (DDR/DDR2/DDR3) DRAMs, which uses a small bridge chip on each DRAM DIMM to break a conventional DRAM rank into multiple smaller mini-ranks so as to reduce the number of devices involved in a single memory access. The design dramatically reduces the memory power consumption with only a slight increase on the memory idle latency. It does not change the DDRx bus protocol and its configuration can be adapted for the best performance-power trade-offs. Our experimental results using four-core multiprogramming workloads show that using x32 mini-ranks reduces memory power by 27.0% with 2.8% performance penalty and using x16 mini-ranks reduces memory power by 44.1% with 7.4% performance penalty on average for memory-intensive workloads, respectively.

Error correcting code predication system and method

Abstract: In memory devices that degrade with use, a memory controller may monitor and record a usage history of portions of the memory. The memory controller can then vary a strength of error correction coding to protect information written to various portions of the memory having different usage histories. More specifically, and memory can receive information to be stored in the memory, select a portion of memory to store the information, and store the information in the selected portion of the memory with an error correction coding having a strength that is based on a usage history of the selected portion of the memory.

Fault-tolerant non-volatile integrated circuit memory

Abstract: Apparatus and methods are disclosed, such as those that store data in a plurality of non-volatile integrated circuit memory devices, such as NAND flash, with convolutional encoding. A relatively high code rate for the convolutional code consumes relatively little extra memory space. In one embodiment, the convolutional code is spread over portions of a plurality of memory devices, rather than being concentrated within a page of a particular memory device. In one embodiment, a code rate of m/n is used, and the convolutional code is stored across n memory devices.

Flash memory device and method for adjusting read voltage of flash memory device

Abstract: A flash memory device includes a cell array and a read voltage adjuster. The cell array includes a first field having first memory cells and a second field having second memory cells. The read voltage adjuster determines a read voltage for reading first data from the first memory cells of the first field with reference to second data read from the memory cells of the second field.

Command interface for memory devices

Abstract: A method for operating a memory device that includes a plurality of analog memory cells includes accepting at an input of the memory device a self-contained command to perform a memory access operation on at least one of the memory cells. The command includes an instruction specifying the memory access operation and one or more parameters that are indicative of analog settings to be applied to the at least one of the memory cells when performing the memory access operation. The self-contained command is executed in the memory device by extracting the parameters, applying the analog settings to the at least one of the memory cells responsively to the extracted parameters, and performing the specified memory access operation in accordance with the instruction on the at least one of the memory cells using the settings.

Technique for memory imprint reliability improvement

Abstract: One embodiment of the present invention relates to a method of reducing imprint of a memory cell. The method comprises adding an inversion condition bit operably associated with one or more memory cells storing a memory word. The inversion condition bit indicates whether the memory word represents an actual payload or an inversion of the actual payload. The inversion condition bit and memory word are selectively toggled by a control circuitry. Inversion is performed by reading the inversion condition bit and memory word and rewriting the memory word back to the one or more memory cells in an inverted or non-inverted state, depending on an inversion condition bit. The inversion condition bit is then written to the inversion status bit value. The memory address is incremented, and the inversion status data state is toggled once the address counter addresses the entire memory array. Other methods and circuits are also disclosed.

Hierarchical immutable content-addressable memory processor

Abstract: Improved memory management is provided according to a Hierarchical Immutable Content Addressable Memory Processor (HICAMP) architecture. In HICAMP, physical memory is organized as two or more physical memory blocks, each physical memory block having a fixed storage capacity. An indication of which of the physical memory blocks is active at any point in time is provided. A memory controller provides a non-duplicating write capability, where data to be written to the physical memory is compared to contents of all active physical memory blocks at the time of writing, to ensure that no two active memory blocks have the same data after completion of the non-duplicating write.

Processor arrays made of standard memory cells

Abstract: Standard memory circuits are used for executing a sum-of-products function between data stored in the memory and data introduced into the memory. The sum-of-products function is executed in a manner substantially similar to a standard memory read operation. The memory circuits are standard or slightly modified SRAM and DRAM cells, or computing memory arrays (CAMs).

Memory circuit arrangement and method for the production thereof

Abstract: A memory circuit arrangement and fabrication method thereof are presented in which the parts of the memory circuit arrangement are situated on two different substrates An integrated memory cell array is situated on one substrate An integrated control circuit that controls access to the memory cells is situated on the other (logic circuit) substrate The control circuit controls sequences when reading, writing or erasing content of a memory cell The logic circuit substrate also contains a CPU and encryption coprocessor The memory circuit contains a sense amplifier, with the aid of which the memory state of a memory cell can be determined, and a decoding circuit that selects a word or bit line

Read circuitry for an integrated circuit having memory cells and/or a memory cell array, and method of operating same

Abstract: An integrated circuit device (e.g., a logic device or a memory device) having a memory cell array which includes (i) a plurality of memory cells, wherein each memory cell is programmable to store one of a plurality of data states, and (ii) a bit line, having a plurality of memory cells coupled thereto. Memory cell control circuitry applies one or more read control signals to perform a read operation wherein, in response to the read control signals, a selected memory cell conducts a current which is representative of the data state stored therein. Sense amplifier circuitry senses the data state stored in the selected memory cell using a signal which is responsive to the current conducted by the selected memory cell. Current regulation circuitry is responsively and electrically coupled to the bit line during a portion of the read operation to sink or source at least a portion of the current provided on the bit line. Sensing circuitry responsively couples the current regulation circuitry to the bit line during the portion of the read operation.

Autonomous refresh of floating body cell (FBC)

Abstract: Physics of autonomous refresh of FBC is presented. Current input to the floating body by impact ionization and output by charge pumping can balance to make FBC refresh by itself without sense amplifier operation. Thanks to this feature, multiple cells on a BL can be refreshed simultaneously, leading to a drastic reduction of BL charging current compared to the conventional refresh. 600 muA refresh current for 1 G-bit memory is achieved in 32 nm technology node with 4 ms retention time. If gate direct tunneling current is used as output, FBC can realize static RAM without periodical refresh when retaining data.

Systems and methods for data transfers between memory cells

Abstract: Systems and methods for reducing the latency of data transfers between memory cells by enabling data to be transferred directly between sense amplifiers in the memory system. In one embodiment, a memory system uses a conventional DRAM memory structure having a pair of first-level sense amplifiers, a second-level sense amplifier and control logic for the sense amplifiers. Each of the sense amplifiers is configured to be selectively coupled to a data line. In a direct data transfer mode, the control logic generates control signals that cause the sense amplifiers to transfer data from a first one of the first-level sense amplifiers (a source sense amplifier) to the second-level sense amplifier, and from there to a second one of the first-level sense amplifiers (a destination sense amplifier.) The structure of these sense amplifiers is conventional, and the operation of the system is enabled by modified control logic.

Data storage in analog memory cell arrays having erase failures

Abstract: A method for data storage includes performing an erasure operation on a group of analog memory cells (32). One or more of the memory cells in the group, which failed the erasure operation, are identified as erase-failed cells. A storage configuration that is used for programming the analog memory cells in the group is modified responsively to the identified erase-failed cells. Data is stored in the group of the analog memory cells using the modified storage configuration.

Nonvolatile semiconductor memory system

Abstract: According to an embodiment, a nonvolatile semiconductor memory system includes: a nonvolatile semiconductor memory; and a memory controller having: a memory interface unit that inputs commands to the nonvolatile semiconductor memory and inputs or outputs data between the nonvolatile semiconductor memory; a memory that stores writing information indicating a memory cell transistor that is written the latest in each of the NAND cell units; and a processor that sets a read voltage based on the writing information to read out data from the memory cell transistors connected to a first word line; wherein a row controller is configured to set a plurality of levels of the read voltage to be applied to the first word line, with respect to one threshold for discriminating data stored in a memory cell transistors.

System and method for on-board diagnostics of memory modules

Abstract: A memory hub includes an on-board diagnostic engine through which diagnostic testing and evaluation of the memory system can be performed. The memory hub includes a link interface for receiving memory requests for access to memory devices of the memory system and a memory device interface coupled to the memory devices for coupling memory requests to the memory devices. A memory hub diagnostic engine is coupled through a switch to the link interface and the memory device interface to perform diagnostic testing of the memory system. The diagnostic engine includes a maintenance port that provides access to results of the diagnostic testing and through which diagnostic testing commands can be received.

Apparatus for detecting and recovering from data destruction caused in an unaccessed memory cell by read, and method therefor

Abstract: Read error in a flash memory destroys data that is not requested to be read, and an efficient read disturb check method is therefore needed. In addition, data may be destroyed beyond repair by error correction before a read error check is run. A non-volatile data storage apparatus including a plurality of memory cells and a memory controller, in which the memory controller is configured to: count how many times data read processing has been executed in memory cells within the management area; read, when the data read processing count that is counted for a first management area exceeds a first threshold, data and an error correction code that are stored in the memory cells within the first management area; decode the read error correction code; and write the data corrected by decoding the error correction code in other management areas than the first management area.

Hybrid memory module

Abstract: One embodiment of the present invention sets forth a hybrid memory module that combines memory devices of different types while presenting a single technology interface. The hybrid memory module includes a number of super-stacks and a first interface configured to transmit data between the super-stacks and a memory controller. Each super-stack includes a number of sub-stacks, a super-controller configured to control the sub-stacks, and a second interface configured to transmit data between the sub-stacks and the first interface. Combining memory devices of different types allows utilizing the favorable properties of each type of the memory devices, while hiding their unfavorable properties from the memory controller.

Low Power Memory Device

Abstract: A method of operation within a memory device is disclosed. The method comprises receiving address information and corresponding enable information in association with a memory access request. The address information includes a row address that specifies a row of storage cells within a storage array of the memory device, and the enable information includes first and second enable values that correspond respectively to first and second storage locations within the row of storage cells. The method involves selectively transferring data between the first and second storage locations and sense amplifier circuitry according to states of the first and second enable values. This includes transferring data between the first storage location and the sense amplifier circuitry if the first enable value is in an enable state and transferring data between the second storage location and the sense amplifier circuitry if the second enable value is in the enable state. The states of the first and second enable values may be separately controlled.

Interface circuit system and method for autonomously performing power management operations in conjunction with a plurality of memory circuits

Abstract: A memory circuit power management system and method are provided. In use, an interface circuit is in communication with a plurality of memory circuits and a system. The interface circuit is operable to interface the memory circuits and the system for autonomously performing a power management operation in association with at least a portion of the memory circuits.

Techniques for improved timing control of memory devices

Abstract: Techniques for improved timing control of memory devices are disclosed. In one embodiment, the techniques may be realized as a memory controller to communicate with a memory device via a communications link. The memory controller may comprise a memory interface to exchange data with the memory device via a set of N conductors according to at least one clock, the data being encoded such that each M bits of data are represented by at least one symbol and each symbol is associated with a combination of signal levels on a group of n conductors, wherein M

Storage system and control method thereof

Abstract: Proposed are a storage system and its control method capable of dealing with the unique problems that arise when using a nonvolatile memory as the memory device while effectively preventing performance deterioration. This storage system is provided with a plurality of memory modules having one or more nonvolatile memory chips, and a controller for controlling the reading and writing of data from and in each memory module. The memory module decides the nonvolatile memory chip to become a copy destination of data stored in the nonvolatile memory when a failure occurs in the nonvolatile memory chip of a self memory module, and copies the data stored in the failed nonvolatile memory chip to the nonvolatile memory chip decided as the copy destination.

System, apparatus, and method for modifying the order of memory accesses

Abstract: Systems and methods for controlling memory access operation are disclosed. The system may include one or more requestors performing requests to memory devices. Within a memory controller, a request queue receives requests from a requestor, a bank decoder determines a destination bank, and the request is placed in an appropriate bank queue. An ordering unit determines if the current request can be reordered relative to the received order and generates a new memory cycle order based on the reordering determination. The reordering may be based on whether there are multiple requests to the same memory page, multiple reads, or multiple writes. A memory interface executes each memory request in the memory cycle order. A data buffer holds write data until it is written to the memory and read data until it is returned to the requestor. The data buffer also may hold memory words used in read-modify-write operations.

Flash memory device capable of storing multi-bit data and single-bit data

Abstract: There is provided a flash memory device capable of manipulating multi-bit and single-bit data. The flash memory device can include a memory cell array with a plurality of memory blocks. The flash memory device can also include a judgment circuit for storing multi-bit/single-bit information indicating whether each of the memory blocks is a multi-bit memory block or not, determining whether or not a memory block of an inputted block address is a multi-bit memory block according to the stored multi-bit/single-bit information and outputting an appropriate flag signal. A read/write circuit for selectively performing multi-bit and single-bit read/program operations of the memory block corresponding to the block address is also included, as well as control logic for controlling the read/write circuit such that the read/write circuit can perform multi-bit or single-bit read/program operations based on the flag signal. An error checking and correction (ECC) circuit including a multi-bit ECC unit and a single-bit ECC unit for checking and correcting an error in a data of the read/write circuit can also be included.

Flash memory device with reduced coupling effect among cells and method of driving the same

Abstract: Embodiments of the invention provide a flash memory device that can improve the reliability of a reading operation by minimizing a variation in the threshold voltage distribution that occurs due to coupling between cells, and a method of driving the flash memory device. In an embodiment of the invention, the method of driving the flash memory includes: performing an erasing operation on memory cells; after the performing the erasing operation, performing a post-programming operation to control a threshold voltage of the memory cells; and after performing the post-programming operation, performing a main programming operation on the memory cells, wherein the performing of the post-programming operation comprises increasing the threshold voltage of the memory cells in an erased state, thereby reducing a difference in the threshold voltage between the memory cells in the erased state and the memory cells in the programmed state.

Computer main memory incorporating volatile and non-volatile memory

Abstract: A main memory for a computer system comprises a controller including an interface to one or more processors, non-volatile memory, and volatile memory. The main memory comprises one or more contiguous range of real addresses supported by both the non-volatile memory and the volatile memory. The controller may be incorporated into a mainboard and the non-volatile memory and the volatile memory may comprise pluggable memory modules. Alternatively, the controller may be incorporated into a hybrid pluggable memory module including non-volatile memory and volatile memory. The controller may utilize the volatile memory as a cache for the non-volatile memory. One or more subsets of the non-volatile memory may be configured to contain a system image, an operating system managed emulated disk image, and/or an operating system managed a page-file. The controller may encrypt and/or compress data written to and/or decrypt and/or decompress data read from the non-volatile memory.

Non-volatile memory and method for predictive programming

Abstract: In a nonvolatile memory having an array of memory cells, wherein the memory cells are individually programmable to one of a range of threshold voltage levels, there is provided a predictive programming mode in which a predetermined function predicts what programming voltage level needs to be applied in order to program a given memory cell to a given target threshold voltage level. In this way, no verify operation needs to be performed, thereby greatly improving the performance of the programming operation. In a preferred embodiment, the predetermined function is linear and is calibrated for each memory cell under programming by one or more checkpoints. The checkpoint is an actual programming voltage that programs the memory cell in question to a verified designated threshold voltage level.

Memory System having Spare Memory Devices Attached to a Local Interface Bus

Abstract: A memory system includes a memory controller, one or more memory channel(s), and a memory subsystem having a memory interface device (e.g. a hub or buffer device) located on a memory subsystem (e.g. a DIMM) coupled to the memory channel to communicate with the memory device(s) array. This buffered DIMM is provided with one or more spare chips on the DIMM, wherein the data bits sourced from the spare chips are connected to the memory hub device and the bus to the DIMM includes only those data bits used for normal operation. The buffered DIMM with one or more spare chips on the DIMM has the spare memory shared among all the ranks, and the memory hub device includes separate control bus(es) for the spare memory device to allow the spare memory device(s) to be utilized to replace one or more failing bits and/or devices within any rank of memory in the memory subsystem.

Memory system, multi-bit flash memory device, and associated methods

Abstract: A memory system includes a multi-bit flash memory device and a flash controller configured to control the multi-bit flash memory device. The flash controller is configured to output a series of commands, pointers, and addresses to the multi-bit flash memory device for read/program operations.

Data conditioning to improve flash memory reliability

Abstract: Methods and apparatus for managing data storage in memory devices utilizing memory arrays of varying density memory cells. Data can be initially stored in lower density memory. Data can be further read, compacted, conditioned and written to higher density memory as background operations. Methods of data conditioning to improve data reliability during storage to higher density memory and methods for managing data across multiple memory arrays are also disclosed.

System and method for error correction and detection in a memory system

Abstract: A system and method for error correction and detection in a memory system. The system includes a memory controller, a plurality of memory modules and a mechanism. The memory modules are in communication with the memory controller and with a plurality of memory devices. The mechanism detects that one of the memory modules has failed possibly coincident with a memory device failure on an other of the memory modules. The mechanism allows the memory system to continue to run unimpaired in the presence of the memory module failure and the memory device failure.