Micron Technology Incorporated

Samir Mittal

Gurpreet Anand

Anirban Ray

Ying Yu Tai

Paul Roger Stonelake

Parag R Maharana

Cheng Yuan Wu

Jiangli Zhu

Wei Wang

Tomoko Ogura Iwasaki

Horia Cristian Simionescu

Edward C McGlaughlin

Manik Advani

Samir Rajadnya

Robert Michael Walker

Robert W Walker

A data bus coupled to a plurality of memory devices is determined to be in a read mode. Responsive to determining that the data bus is in the read mode, a particular read operation identified in a particular memory queue of memory queues that include identifiers of one or more write operations and identifiers of one or more read operations is determined. The particular memory queue includes a highest number of read operations for a memory device of the memory devices. The particular read operation is transmitted from the particular memory queue over the data bus.

Scheduling of read operations and write operations based on a data bus mode

A computing system having a memory component with an embedded media controller. The memory component is encapsulated within an integrated circuit (IC) package. The embedded controller within the IC package is configured to: receive incoming packets, via a serial communication interface of the controller, from a serial connection outside of the IC package; convert the incoming packets into commands and addresses according to a predetermined serial communication protocol; operate memory units encapsulated within the IC package according to the commands and the addresses; convert results of at least a portion of the commands into outgoing packets; and transmit the outgoing packets via the serial communication interface to the serial connection outside of the IC package.

Memory systems having controllers embedded in packages of integrated circuit memory

A computing system having a power loss detector and memory components to store data associated with write commands received from a host system. The write commands are flushed from a protected write queue of the host system responsive to detecting an impending loss of power. The computing system further includes a processing device to receive the write commands over a memory interface. The processing device is further to, responsive to detecting the loss of power by the detector: disable the memory interface, and store the data associated with write commands that are received prior to disabling the memory interface. The data is stored in one or more of the memory components using power supplied by one or more capacitors.

CPU cache flushing to persistent memory

An access request is received. The access request comprises a physical page address corresponding to a primary memory block of a memory device, an input security key, and a logical page address corresponding to the physical page address. The input security key is provided as input to a (CAM) block that stores a plurality of security keys to verify that the input security key matches a stored security key. A location of the stored security key is checked to verify that it corresponds to the logical page address included in the access request based a predetermined mapping. Based on verifying that the stored security key corresponds to the logical page address included in the access request, the physical page address corresponding to the primary memory block is accessed.

Memory device data security based on content-addressable memory architecture

A memory system having memory components, a remote direct memory access (RDMA) network interface card (RNIC), and a host system, and configured to: allocate a page of virtual memory for an application; map the page of virtual memory to a page of physical memory in the memory components; instruct the RNIC to perform an RDMA operation; perform, during the RDMA operation, a data transfer between the page of physical memory in the plurality of memory components and a remote device that is connected via a computer network to the remote direct memory access network interface card; and at least for a duration of the data transfer, lock a mapping between the page of virtual memory and the page of physical memory in the memory components.

Remote direct memory access in multi-tier memory systems

A data bus is determined to be in a write mode. Whether a number of memory queues that identify at least one write operation satisfies a threshold criterion is determined. The memory queues include identifiers of one or more write operations and identifiers of one or more read operations. Responsive to determining that the number of memory queues satisfies the threshold criterion, a write operation from the memory queues is transmitted over the data bus.

A sequencer component residing in a first package receives data from a controller residing in a second package that is different than the first package including the sequencer component. The sequencer component performs an error correction operation on the data received from the controller. The error correction operation encodes the data with additional data to generate a code word. The sequencer component stores the code word at a memory device.

Memory sub-system including an in-package sequencer to perform error correction and memory testing operations

A processing device, operatively coupled with one or more memory devices, is configured to provide a plurality of virtual memory controllers, partition one or more memory devices into a plurality of physical partitions, and associate each of the plurality of virtual memory controllers with one of the plurality of physical partitions. The processing device further provides a plurality of physical functions, wherein each of the plurality of physical functions corresponds to a different one of the plurality of virtual memory controllers, and presents the plurality of physical functions to a host computing system over a peripheral component interconnect express (PCIe) interface.

NVMe direct virtualization with configurable storage

An instruction can be received at a sequencer from a controller. The sequencer can be in a package including the sequencer and one or more memory components. The sequencer is operatively coupled to a controller that is separate from the package. A processing device of the sequencer can perform an operation based on the instruction on at least one of the one or more memory components in the package.

Memory sub-system including an in package sequencer separate from a controller

A computing system having memory components of different tiers. The computing system further includes a controller, operatively coupled between a processing device and the memory components, to: receive from the processing device first data access requests that cause first data movements across the tiers in the memory components; service the first data access requests after the first data movements; predict, by applying data usage information received from the processing device in a prediction model trained via machine learning, second data movements across the tiers in the memory components; and perform the second data movements before receiving second data access requests, where the second data movements reduce third data movements across the tiers caused by the second data access requests.

Predictive data orchestration in multi-tier memory systems

A processing device in a host system monitors a data temperature of a plurality of memory pages stored in a host-addressable region of a cache memory component operatively coupled with the host system. The processing device determines that a first memory page of the plurality of memory pages satisfies a first threshold criterion pertaining to the data temperature of the first memory page and sends a first migration command indicating the first memory page to a direct memory access (DMA) engine executing on a memory-mapped storage component operatively coupled with the cache memory component and with the memory-mapped storage component via a peripheral component interconnect express (PCIe) bus. The first migration command causes the DMA engine to initiate a first DMA transfer of the first memory page from the cache memory component to a host-addressable region of the memory-mapped storage component.

Memory tiering using PCIe connected far memory

A graph can be generated based on an access pattern associated with blocks of a memory device that have been accessed by a host system, wherein the graph comprises nodes representing at least a subset of the blocks that have been accessed by the host system and edges that are based on the access pattern, wherein each edge is associated with a respective probability value between a respective pair of nodes. A number of edges having respective probability values that satisfy a probability value threshold criterion can be determined. It can be determined whether the number of edges satisfies a decayed edge value condition. In response to determining that the number of edges does not satisfy the decayed edge value condition, the graph can be removed.

Storing data based on a probability of a data graph

A computing system having memory components, including first memory and second memory. The computing system further includes a processing device, operatively coupled with the memory components, to: receive, in a prediction engine, usage history of pages in the second memory; train a prediction model based on the usage history; predict, by the prediction engine using the prediction model, likelihood of the pages being used in a subsequent period of time; and responsive to the likelihood predicted by the prediction engine, copy by a controller data in a page in the second memory to the first memory.

Predictive paging to accelerate memory access

A processing device can determine a configuration parameter based on a memory type of a memory component that is managed by a memory system controller. The processing device can receive data from a host system. The processing device can generate, by performing a memory operation using the configuration parameter, an instruction based on the data. The processing device can identify a sequencer of a plurality of sequencers that are collocated, within a single package external to the memory system controller, wherein each sequencer of the plurality of sequencers interfaces with a respective memory component. The processing device can send the instruction to the sequencer.

Growing community of inventors

Palo Alto, CA, United States of America

Samir Mittal