# US 2025/0165193 A1 · Remote direct memory access in multi-tier memory systems

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0165193 A1
- Document ID: US20250165193A1
- Status: Published application
- Kind: A1
- Application number: 19/023,089
- Filed: 2025-01-15
- Published: 2025-05-22
- Earliest claimed: 2018-02-05
- Assignees: Micron Technology, Inc.
- Applicants: Micron Technology, Inc.
- Inventors: Parag R. Maharana; Anirban Ray; Gurpreet Anand; Samir Mittal
- Web page: https://patentgrep.com/patent/US20250165193A1

## CPC

- G06F 3/067: {Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]}
- G06F 12/0868: Data transfer between cache memory and other subsystems, e.g. storage devices or host systems
- G06F 12/00: Accessing, addressing or allocating within memory systems or architectures (digital input from, or digital output to record carriers, e.g. to disk storage units, G06F3/06)
- G06F 12/1081: for peripheral access to main memory, e.g. direct memory access [DMA]
- G06F 12/126: with special data handling, e.g. priority of data or instructions, handling errors or pinning
- G06F 13/28: using burst mode transfer, e.g. direct memory access {DMA}, cycle steal (G06F13/32 takes precedence)
- G06F 15/17331: {Distributed shared memory [DSM], e.g. remote direct memory access [RDMA]}
- G06N 3/09: Supervised learning
- G06N 3/092: Reinforcement learning
- G11C 14/0009: {in which the volatile element is a DRAM cell}
- H04L 67/1097: for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
- G06F 2212/1024: Latency reduction
- G06F 2212/154: Networked environment
- G06F 2212/502: using adaptive policy
- G06F 2212/657: Virtual address space management
- G06F 2212/7201: Logical to physical mapping or translation of blocks or pages
- G06N 3/065: Analogue means
- G06N 3/08: Learning methods
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G11C: STATIC STORES (semiconductor memory devices H10B)
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

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.

## Claims

### Claim 1

A device, comprising:
remote direct memory access network interface card configured to transfer data between a remote apparatus connected to the computer network and a memory module configured on a memory bus in response to a command received via a peripheral interconnect.

### Claim 2 (depends on 1)

The device of claim 1, further comprising:
the memory bus; and
a memory module connected to the memory bus.

### Claim 3 (depends on 2)

The device of claim 2, further comprising:
the peripheral interconnect; and
a host system connected to the memory bus and connected to the peripheral interconnect to send the command.

### Claim 4 (depends on 3)

The device of claim 3, wherein the host system is configured to run a driver programmed for the remote direct memory access network interface card and configured to generate the command.

### Claim 5 (depends on 4)

The device of claim 4, wherein the remote direct memory access network interface card is configured to implement single root input/output virtualization to provide a physical function and a plurality of virtual functions.

### Claim 6 (depends on 5)

The device of claim 5, wherein the driver running in a hypervisor in the host system is configured to control access to the physical function; and a driver running in a virtual machine in the host system is configured to control access to one of the plurality of virtual functions.

### Claim 7 (depends on 6)

The device of claim 6, wherein the memory module includes a first memory and a second memory slower than the first memory; and the remote direct memory access network interface card is configured to transfer the data of a virtual memory page between the remote apparatus and the second memory in the memory module.

### Claim 8 (depends on 7)

The device of claim 7, wherein the host system is configured to:
map the virtual memory page to the second memory before the remote direct memory access network interface card transferring the data between the remote apparatus and the second memory in the memory module; and
prevent the virtual memory page from being mapped to the first memory during the remote direct memory access network interface card transferring the data between the remote apparatus and the second memory in the memory module.

### Claim 9 (depends on 8)

The device of claim 8, wherein the driver running in the hypervisor in the host system is configured to detect a completion of the remote direct memory access network interface card transferring the data between the remote apparatus and the second memory in the memory module; and the host system is configured to allow the virtual memory page to be mapped to the first memory after the completion.

### Claim 10 (depends on 9)

The device of claim 9, wherein the first memory is a volatile memory and the second memory is a non-volatile memory.

### Claim 11

A method, comprising:
receiving, in a remote direct memory access network interface card via a peripheral interconnect, a command; and
transferring, in response to the command, data between a remote apparatus connected to a computer network and a memory module configured on a memory bus.

### Claim 12 (depends on 11)

The method of claim 11, wherein a host system connected to the memory bus and connected to the peripheral interconnect is configured to send the command.

### Claim 13 (depends on 12)

The method of claim 12, further comprising:
running, in the host system, a driver programmed for the remote direct memory access network interface card and configured to generate the command.

### Claim 14 (depends on 13)

The method of claim 13, wherein the memory module includes a first memory and a second memory slower than the first memory; and the remote direct memory access network interface card is configured to transfer the data of a virtual memory page between the remote apparatus and the second memory in the memory module.

### Claim 15 (depends on 14)

The method of claim 14, further comprising:
mapping the virtual memory page to the second memory before the remote direct memory access network interface card transferring the data between the remote apparatus and the second memory in the memory module; and
preventing the virtual memory page from being mapped to the first memory during the remote direct memory access network interface card transferring the data between the remote apparatus and the second memory in the memory module.

### Claim 16 (depends on 15)

The method of claim 15, further comprising:
detecting, by the driver running in the host system, a completion of the remote direct memory access network interface card transferring the data between the remote apparatus and the second memory in the memory module; and
allowing, after the completion, the virtual memory page to be mapped to the first memory.

### Claim 17 (depends on 16)

The method of claim 16, wherein the first memory is a volatile memory and the second memory is a non-volatile memory.

### Claim 18

A non-transitory computer storage medium storing instructions which, when executed by a computing system, cause the computing system to perform a method, the method comprising:
generating a command to a remote direct memory access network interface card having:
before sending the command to the remote direct memory access network interface card to cause the remote direct memory access network interface card to transfer data of a virtual memory page between a remote apparatus connected to the computer network and a memory module configured on the memory bus:
preventing the virtual memory page from being mapped to first memory of the memory module for duration of the remote direct memory access network interface card transferring the data between the remote apparatus and the memory module.

### Claim 19 (depends on 18)

The non-transitory computer storage medium of claim 18, wherein the method further comprises:
detecting a completion of the remote direct memory access network interface card transferring the data between the remote apparatus and the second memory in the memory module; and
allowing, after the completion, the virtual memory page to be mapped to the first memory.

### Claim 20 (depends on 18)

The non-transitory computer storage medium of claim 18, wherein the first memory is a volatile memory and the second memory is a non-volatile memory.
