# US 2025/0061066 A1 · Memory appliance couplings and operations

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

- Number: US 2025/0061066 A1
- Document ID: US20250061066A1
- Status: Published application
- Kind: A1
- Application number: 18/817,173
- Filed: 2024-08-27
- Published: 2025-02-20
- Earliest claimed: 2015-04-10
- Assignees: Rambus Inc.
- Applicants: Rambus Inc.
- Inventors: Vlad Fruchter; Keith Lowery; George Michael Uhler; Steven Woo; Chi-Ming (Philip) Yeung; Ronald Lee
- Web page: https://patentgrep.com/patent/US20250061066A1

## CPC

- G06F 13/1668: {Details of memory controller}
- G06F 13/4022: {using switching circuits, e.g. switching matrix, connection or expansion network (G06F13/4009 takes precedence)}
- G06F 13/4068: {Electrical coupling}
- G06F 13/4282: {on a serial bus, e.g. I2C bus, SPI bus (on daisy chain buses G06F13/4247)}
- Y02D 10/00: Energy efficient computing, e.g. low power processors, power management or thermal management
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- Y02D: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE

## Abstract

System and method for improved transferring of data involving memory device systems. A memory appliance (MA) comprising a plurality of memory modules is configured to store data within the plurality of memory modules and further configured to receive data commands from the first server and a second server coupled to the MA. The data commands may include direction memory access commands such that the MA can service the data commands while bypassing a host controller of the MA.

## Claims

### Claim 1

(canceled)

### Claim 2

A memory appliance (MA) comprising:
a central processing unit (CPU);
a plurality of memory modules, wherein the MA is configured to store data within the plurality of memory modules; and
a local interconnect coupled to the plurality of memory modules, wherein at least one server of a plurality of servers is directly coupled to the local interconnect through a direct connection, and wherein the local interconnect comprises a translation engine to translate a memory address associated with a request received via the direct connection to bypass the CPU of the MA and access the data stored in the plurality of memory modules responsive to the request.

### Claim 3 (depends on 2)

The memory appliance of claim 2, wherein the plurality of servers is connected to the MA via a network switch.

### Claim 4 (depends on 3)

The memory appliance of claim 3, wherein the direct connection bypasses the network switch.

### Claim 5 (depends on 2)

The memory appliance of claim 2, wherein the direct connection comprises a PERIPHERAL COMPONENT INTERCONNECT EXPRESS (PCle)® connection.

### Claim 6 (depends on 2)

The memory appliance of claim 2, wherein the request comprises a direct memory access (DMA) request.

### Claim 7 (depends on 2)

The memory appliance of claim 2, wherein the local interconnect is to:
receive the request via the direct connection;
translate the memory address associated with the request to generate a translated address associated with one of the plurality of memory modules of the MA; and
send an access request to the one of the plurality of memory modules, the access request comprising the translated address.

### Claim 8 (depends on 7)

The memory appliance of claim 7, wherein the local interconnect is further to:
receive the data from the one of the plurality of memory modules; and
send the data to the at least one server of the plurality of servers via the direct connection.

### Claim 9

A system comprising:
a network switch;
a memory appliance (MA) coupled to the network switch, the MA comprising a plurality of memory modules and a local interconnect, wherein the MA is configured to store data within the plurality of memory modules; and
a plurality of servers coupled to the network switch, wherein at least one server of the plurality of servers is directly coupled to the local interconnect of the MA through a direct connection;
wherein the local interconnect of the MA comprises a translation engine to translate a memory address associated with a request received via the direct connection to bypass a host central processing unit (CPU) of the MA and access the data stored in the MA responsive to the request.

### Claim 10 (depends on 9)

The system of claim 9, wherein the network switch is external to the MA.

### Claim 11 (depends on 10)

The system of claim 10, wherein the direct connection bypasses the network switch.

### Claim 12 (depends on 9)

The system of claim 9, wherein the direct connection comprises a PERIPHERAL COMPONENT INTERCONNECT EXPRESS (PCle)® connection.

### Claim 13 (depends on 9)

The system of claim 9, wherein the request comprises a direct memory access (DMA) request.

### Claim 14 (depends on 9)

The system of claim 9, wherein the local interconnect is to:
receive the request via the direct connection;
translate the memory address associated with the request to generate a translated address associated with one of the plurality of memory modules of the MA; and
send an access request to the one of the plurality of memory modules, the access request comprising the translated address.

### Claim 15 (depends on 14)

The system of claim 14, wherein the local interconnect is further to:
receive the data from the one of the plurality of memory modules; and
send the data to the at least one server of the plurality of servers via the direct connection.

### Claim 16

A method comprising:
receiving, from at least one server of a plurality of servers, a request having an associated memory address for a memory appliance (MA) coupled to a network switch, the MA comprising a plurality of memory modules and a local interconnect, wherein the plurality of servers are coupled to the network switch, and wherein the at least one server is directly coupled to the local interconnect of the MA through a direct connection;
translating, by a translation engine of the local interconnect in the MA, the associated memory address to a translated memory address; and
accessing data stored in the plurality of memory modules at the translated memory address responsive to the request.

### Claim 17 (depends on 16)

The method of claim 16, wherein the network switch is external to the MA, and wherein the direct connection bypasses the network switch.

### Claim 18 (depends on 16)

The method of claim 16, wherein the local interconnect comprises a PERIPHERAL COMPONENT INTERCONNECT EXPRESS (PCIe)® interconnect.

### Claim 19 (depends on 16)

The method of claim 16, wherein the request comprises a direct memory access (DMA) request.

### Claim 20 (depends on 16)

The method of claim 16, further comprising:
receiving, by the local interconnect, the request via the direct connection;
translating, by the translation engine of the local interconnect, the associated memory address to generate the translated address associated with one of the plurality of memory modules of the MA; and
sending, by the local interconnect, an access request to the one of the plurality of memory modules, the access request comprising the translated address.

### Claim 21 (depends on 20)

The method of claim 20, further comprising:
receiving, by the local interconnect, the data from the one of the plurality of memory modules; and
sending, by the local interconnect, the data to the at least one server of the plurality of servers via the direct connection.
