Intel NDTM US LLC

Methods and apparatus for fast and efficient verify recovery and array discharge for 3D NAND memory arrays and other 3D storage devices. The 3D storage device includes storage arrays including strings of memory cells stacked on top of one another and sharing a channel in a pillar for the string. The memory cells for a string occupy respective tiers in a 3D structure with each tier having an associated wordline. A controller is used to program charge levels in the memory cells. Programming is followed by a fast verify recovery where a voltage is applied to the wordlines to perform a program verify, followed by discharging wordlines. Erased wordlines are identified and discharged first, followed by programmed wordlines, which may employ staggered discharge sequences. Dummy wordlines are then discharged, with an optional timer delay. For multi-deck devices, wordlines in the deck with an active wordline are discharged before wordlines in one or more other decks.

An embodiment of an electronic apparatus may include one or more substrates, and logic coupled to the one or more substrates, the logic to control access to a persistent storage media based on a block and sub-block access structure, store a data structure in the persistent storage media to track read fails at a sub-block granularity for a word-line for every block, and update the data structure in response to a read fail on a block to indicate a failed sub-block that corresponds to the read fail for a word-line for the block. Other embodiments are disclosed and claimed.

Systems, apparatuses and methods may provide for technology that issues a program pulse to a selected subblock of a NAND memory array, conducts a pulse recovery phase after the program pulse, and shuts down unselected subblocks in the NAND memory array during the pulse recovery phase.

A request to read data stored in a non-volatile memory (NVM) is processed by incrementing a global read counter for the NVM, incrementing a local read counter for a zone of the NVM being accessed by processing of the read request, computing a degree of read hotness for the zone, computing a read concentration of the zone based at least in part on the degree of read hotness of the zone, the global read counter, and the local read counter, and relocating the data in the NVM when the read concentration of the zone meets or exceeds a threshold.

Systems, apparatuses and methods may provide for technology that generates address information for a plurality of planes in NAND memory, excludes column information from the address information, and sends a read command sequence to the NAND memory, wherein the read command sequence includes the address information. In one example, the technology also excludes plane confirm commands and busy cycles from the read command sequence.

A method is described. The method includes programming a column of flash storage cells in a direction along the column in which a parasitic transistor that resides between a cell being programmed and an immediately next cell to be programmed has lower resistivity as compared to a corresponding parasitic transistor that exists if the programming were to be performed in an opposite direction along the column.

A storage device includes a storage array with multiple planes organized as plane groups, where the planes of a plane group receive and process commands in parallel. The storage device includes a storage controller that receives a command from a host controller. In response to receipt of the command the storage controller provides ready information for all planes to the host controller. The multiple planes can optionally have independent multiplane read operation (IMPRO). Each plane group can have a first plane and a second plane, and the storage controller can optionally read data from the first plane of a plane group in response to a virtual ready signal for the first plane, before the second plane of the plane group is ready.

Proactively adjusting read voltages at the system level, before performing a read operation on data located in a partially-programmed block in a block-addressable non-volatile memory, can significantly reduce the re-read trigger rate. This reduces the rate of entering a read recovery flow and subsequent read latency. Determining in advance a wordline-specific pattern of wordline offsets associated with past unsuccessful reads in partially-programmed blocks allows read voltages to be proactively adjusted for vulnerable wordlines. Read voltages are restored for subsequent read operations.

A request to read data stored in a non-volatile memory (NVM) is processed by incrementing a global read counter for the NVM, incrementing a local read counter for a zone of the NVM being accessed by processing of the read request, computing a degree of read hotness for the zone, computing a read concentration of the zone based at least in part on the degree of read hotness of the zone, the global read counter, and the local read counter, and relocating the data in the NVM when the read concentration of the zone meets or exceeds a threshold.

A method is described. The method includes performing the following on a flash memory chip: measuring a temperature of the flash memory chip; and, changing a program step size voltage of the flash memory chip because the temperature of the flash memory chip has changed.