# US 12,216,575 B2 · Patterned memory-network data transfer > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,216,575 B2 - Document ID: US12216575B2 - Status: Granted - Kind: B2 - Application number: 17/858,104 - Filed: 2022-07-06 - Published: 2024-01-11 - Granted: 2025-02-04 - Earliest claimed: 2022-07-06 - Assignees: Mellanox Technologies, Ltd. - Applicants: Mellanox Technologies, Ltd. - Inventors: Daniel Marcovitch; Gil Bloch; Richard Graham; Yossef Itigin; Ortal Ben Moshe; Roman Nudelman - Web page: https://patentgrep.com/patent/US12216575B2 ## CPC - G06F 12/0653: {with centralised address assignment} - G06F 12/0607: {Interleaved addressing} - G06F 12/0835: {for main memory peripheral accesses (e.g. I/O or DMA)} - G06F 12/109: for multiple virtual address spaces, e.g. segmentation (G06F12/1036 takes precedence) - G06F 13/28: using burst mode transfer, e.g. direct memory access {DMA}, cycle steal (G06F13/32 takes precedence) - G06F 13/4027: {using bus bridges (G06F13/4022 takes precedence)} - G06F 13/404: {with address mapping} - G06F 15/17331: {Distributed shared memory [DSM], e.g. remote direct memory access [RDMA]} - H04L 67/141: Setup of application sessions (admission control or resource allocation in data switching networks H04L47/70) - G06F 2212/1016: Performance improvement - G06F 2212/154: Networked environment - G06F 2212/657: Virtual address space management - H04L 67/10: in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) ## Abstract A network device includes a first interface, a second interface, and circuitry. The first interface is configured to communicate at least with a memory. The second interface is configured to communicate over a network with a peer network device. The circuitry is configured to receive a request to transfer data over the network between the memory and the peer network device in accordance with (i) a pattern of offsets to be accessed in the memory and (ii) a memory key representing a memory space to be accessed using the pattern, and to transfer the data in accordance with the request. ## Claims ### Claim 1 A network device, comprising: a first interface, configured to communicate at least with a memory; a second interface, configured to communicate over a network with a peer network device; and circuitry, configured to pre-register a plurality of patterns of offsets, to pre-associate a memory key representing a memory space to be accessed with a given pattern from the plurality of patterns of offsets, to receive a request to transfer data over the network between the memory and the peer network device, wherein the request indicates the memory key, but not the given pattern pre-associated with the memory key, and to transfer the data in accordance with the request using the memory key and the given pattern pre-associated with the memory key. ### Claim 2 (depends on 1) The network device according to claim 1, wherein the circuitry is configured to receive the request as a work-request via the first interface. ### Claim 3 (depends on 1) The network device according to claim 1, wherein each of the pre-registered patterns of offsets comprises one of (i) a pattern that accesses a contiguous memory space, (ii) a pattern that accesses a non-contiguous memory space, and (iii) a strided pattern. ### Claim 4 (depends on 1) The network device according to claim 1, wherein the request specifies multiple memory keys representing respective memory spaces to be accessed using the pattern. ### Claim 5 (depends on 1) The network device according to claim 1, wherein the memory space represented by the memory key is one of a virtual memory space and a physical memory space. ### Claim 6 (depends on 1) The network device according to claim 1, wherein, in response to receiving a re-association instruction, the circuitry is configured to associate the memory key with a different pattern. ### Claim 7 (depends on 1) The network device according to claim 1, wherein the pattern is pre-associated in the network device with at least one additional memory key. ### Claim 8 (depends on 1) The network device according to claim 1, wherein the request specifies a starting virtual address or initial offset parameter for the pattern. ### Claim 9 (depends on 1) The network device according to claim 1, wherein the pattern specifies the offsets using a nested representation that comprises at least one internal pattern. ### Claim 10 (depends on 1) The network device according to claim 1, wherein the pattern specifies the offsets using an iterative representation that defines (i) a basic pattern to be iterated multiple times, (ii) a number of times the basic pattern is to be iterated, and (iii) an offset between successive iterations of the basic pattern. ### Claim 11 (depends on 1) The network device according to claim 1, wherein the request is associated with a scatter or gather command. ### Claim 12 (depends on 1) The network device according to claim 1, wherein the request is associated with an InfiniBand transaction. ### Claim 13 (depends on 1) The network device according to claim 1, wherein the request is associated with an Ethernet transaction. ### Claim 14 A method for data transfer, comprising: in a network device, communicating with a memory and, over a network, with a peer network device; pre-registering, in the network device, a plurality of patterns of offsets; pre-associating a memory key representing a memory space to be accessed with a given pattern from the plurality of patterns of offsets; receiving in the network device a request to transfer data over the network between the memory and the peer network device, wherein the request indicates the memory key, but not the given pattern pre-associated with the memory key; and transferring the data in accordance with the request using the memory key and the given pattern pre-associated with the memory key. ### Claim 15 (depends on 14) The method according to claim 14, wherein receiving the request comprises receiving the request as a work-request via the first interface. ### Claim 16 (depends on 14) The method according to claim 14, wherein each of the multiple pre-registered patterns of offsets comprises one of (i) a pattern that accesses a contiguous memory space, (ii) a pattern that accesses a non-contiguous memory space, and (iii) a strided pattern. ### Claim 17 (depends on 14) The method according to claim 14, wherein the request specifies multiple memory keys representing respective memory spaces to be accessed using the pattern. ### Claim 18 (depends on 14) The method according to claim 14, wherein the memory space represented by the memory key is one of a virtual memory space and a physical memory space. ### Claim 19 (depends on 14) The method according to claim 14, and comprising, in response to receiving a re-association instruction, associating the memory key with a different pattern. ### Claim 20 (depends on 14) The method according to claim 14, wherein the pattern is pre-associated in the network device with at least one additional memory key. ### Claim 21 (depends on 14) The method according to claim 14, wherein the request specifies a starting virtual address or initial offset parameter for the pattern. ### Claim 22 (depends on 14) The method according to claim 14, wherein the pattern specifies the offsets using a nested representation that comprises at least one internal pattern. ### Claim 23 (depends on 14) The method according to claim 14, wherein the pattern specifies the offsets using an iterative representation that defines (i) a basic pattern to be iterated multiple times, (ii) a number of times the basic pattern is to be iterated, and (iii) an offset between successive iterations of the basic pattern. ### Claim 24 (depends on 14) The method according to claim 14, wherein the request is associated with a scatter or gather command. ### Claim 25 (depends on 14) The method according to claim 14, wherein the request is associated with an InfiniBand transaction. ### Claim 26 (depends on 14) The method according to claim 14, wherein the request is associated with an Ethernet transaction.