Search

US-20260129020-A1 - REUSEABLE VIRTUAL NETWORK ADDRESSES

US20260129020A1US 20260129020 A1US20260129020 A1US 20260129020A1US-20260129020-A1

Abstract

Systems and methods herein are for one or more processing units that can communicate in a network using a subnet manager (SM) that includes a mapping of one virtual network address to two or more physical ports, where the one virtual network address may be associated with different switches of different data planes, and where data may be communicated concurrently and separately using the one virtual network address and using the two or more physical ports.

Inventors

  • Lior Hodaya Bezen
  • Alex Netes
  • Idan Seider
  • Lion Levi
  • Glenn Dearth
  • Nitin Hande
  • Mark Hummel

Assignees

  • MELLANOX TECHNOLOGIES, LTD.

Dates

Publication Date
20260507
Application Date
20260105

Claims (20)

  1. 1 . A system for communication in a network, comprising: a subnet manager (SM) to comprise one or more mapping tables of one virtual network address assigned to two or more physical ports, wherein the one virtual network address is associated with different switches which represent different subnets, and wherein data associated with the two or more physical ports is communicated concurrently and separately using the one virtual network address and using the two or more physical ports.
  2. 2 . The system of claim 1 , wherein the different subnets provide physical separation for traffic associated with the two or more physical ports.
  3. 3 . The system of claim 1 , wherein the SM is further to: determine connected devices in the network; determine that the one virtual network address is associated with a first physical port of the two or more physical ports on a host machine and is associated with a first one of the different switches; and assign the one virtual network address to a second physical port of the two or more physical ports on the host machine and to a second one of the different switches.
  4. 4 . The system of claim 1 , wherein the SM is further to: allow re-use of the one virtual network address for the different subnets by an update to the one or more mapping tables to associate the one virtual network address to different ones of the two or more physical ports and to the different switches.
  5. 5 . The system of claim 1 , wherein the SM is further to: use a combination of one of the different subnets and the one virtual network address to uniquely identify one of the two or more physical ports on a host machine for traffic communicated through the different switches.
  6. 6 . The system of claim 1 , wherein the SM is further to: provide first configuration information to a first switch of the different switches to identify the one virtual network address and a first one of the two or more physical ports; and provide second configuration information to a second switch of the different switches to identify the one virtual network address and a second one of the two or more physical ports.
  7. 7 . The system of claim 1 , wherein the SM is further to: determine that a first number of physical ports on connected host machines in the network exceed a second number associated with available virtual network addresses; prepare the different subnets for an assignment; and enable the available virtual network addresses to be assigned, as part of the assignment, to the first number of physical ports based at least on the different subnets available in the network.
  8. 8 . A system for communication in a network, comprising: at least one switch to be part of a first subnet and to receive configuration information from a subnet manager (SM), the configuration information associated with a mapping table in the SM comprising a mapping of one virtual network address assigned to two or more physical ports and to enable the at least one switch to communicate, concurrently and separately, with a further switch of a second subnet using the two or more physical ports that are represented by the one virtual network address.
  9. 9 . The system of claim 8 , wherein the SM is further to: receive a request for communication from a sender host machine and specifying a receiver host machine; and associate the one virtual network address with a first one of the two or more physical ports of the sender host machine and with the first subnet, the one virtual network address associated with a second one of the two or more physical ports of the sender host machine and with the second subnet to enable the sender host machine to perform the communication with the receiver host machine.
  10. 10 . The system of claim 8 , wherein the SM is further to: allow re-use of the one virtual network address by an update to the mapping table to assign the one virtual network address to different ones of the two or more physical ports on a host machine and to further switches of further subnets.
  11. 11 . The system of claim 8 , wherein the SM is further to: provide the configuration information to the at least one switch to identify the one virtual network address and a first one of the two or more physical ports on a host machine; and provide second configuration information to the further switch to identify the one virtual network address and a second one of the two or more physical ports on the host machine.
  12. 12 . The system of claim 8 , wherein the SM is further to: determine that a first number of physical ports on connected host machines in the network exceed a second number associated with available virtual network addresses; prepare the first subnet and the second subnet for an assignment; and enable the available virtual network addresses to be assigned, as part of the assignment, to the first number of physical ports based at least on the first subnet and the second subnet.
  13. 13 . A system comprising: one or more circuits to communicate configuration information between a subnet manager (SM) and a plurality of switches to enable the plurality of switches to provide communication between the plurality of switches using a mapping table of one virtual network address assigned to two or more physical ports that are associated with the plurality of switches, wherein the one virtual network address is associated with different switches of the plurality of switches that are in different subnets, and wherein the communication occurs concurrently and separately using the one virtual network address and using the two or more physical ports.
  14. 14 . The system of claim 13 , wherein the SM is further to: receive a request for communication from a sender host machine and specifying a receiver host machine; and associate the one virtual network address with a first one of the two or more physical ports of the sender host machine and with a first subnet of the different subnets, and the one virtual network address with a second one of the two or more physical ports of the sender host machine and with a second subnet of the different subnets to enable the sender host machine to perform the communication with the receiver host machine.
  15. 15 . The system of claim 13 , wherein the SM is further to: allow re-use of the one virtual network address by an update to the mapping table to assign the one virtual network address to different ones of the two or more physical ports and to further ones of the plurality of switches of further subnets.
  16. 16 . The system of claim 13 , wherein the SM is further to: provide the configuration information to a first switch of the plurality of switches to identify the one virtual network address and a first one of the two or more physical ports on at least one host machine; and provide second configuration information to a second switch of the plurality of switches to identify the one virtual network address and a second one of the two or more physical ports on the at least one host machine.
  17. 17 . The system of claim 13 , wherein the SM is further to: determine that a first number of physical ports on connected host machines in a network exceed a second number associated with available virtual network addresses; prepare the different subnets for an assignment; and enable the available virtual network addresses to be assigned, as part of the assignment, to the first number of physical ports based at least on the different subnets available in the network.
  18. 18 . A method for addressing in a network, the method comprising: providing, in a subnet manager (SM), a mapping table of one virtual network address assigned to two or more physical ports; enabling the one virtual network address to be associated with different switches of different subnets; and communicating, concurrently and separately, data associated with the two or more physical ports using the one virtual network address and using the two or more physical ports.
  19. 19 . The method of claim 18 , wherein the different subnets provide physical separation for traffic associated with the two or more physical ports.
  20. 20 . The method of claim 18 , further comprising: determining connected devices in the network; determining that the one virtual network address is associated with a first physical port of the two or more physical ports on a host machine and is associated with a first one of the different switches; and assigning the one virtual network address to a second physical port of the two or more physical ports on the host machine and to a second one of the different switches.

Description

CROSS-REFERENCE TO RELATED APPLICATION(S) This Continuation Patent Application claims priority to U.S. patent application Ser. No. 18/382,810, filed Oct. 23, 2023, entitled “DATA PLANE SEPARATION IN NETWORKS TO REUSE VIRTUAL NETWORK ADDRESSES,” the full disclosure of which is hereby incorporated by reference herein in its entirety for all intents and purposes. TECHNICAL FIELD At least one embodiment pertains to data plane separation in networks to reuse virtual network addresses. BACKGROUND InfiniBand® (IB) networks may be used in high-performance computing and data centers. IB networks may be a switched fabric network that uses a high-bandwidth and low-latency communication protocol referred to as Remote Direct Memory Access (RDMA) to allow devices to communicate directly without a host. This allows for very fast communication speeds and low latency, making it well-suited for applications that require high-speed data transfer and low-latency communication, such as financial trading systems, high-performance computing, and other time-sensitive applications. In IB networks, communication occurs using a series of interconnected switches, which are responsible for routing data packets between devices. Distinct identifiers used to identify IB ports can limit a number host machines able to communicate on an IB network but also require assignment of the identifier by a number of endpoints or a number of ports per endpoint. BRIEF DESCRIPTION OF DRAWINGS FIG. 1 illustrates a system that is subject to data plane separation in networks to reuse virtual network addresses, according to at least one embodiment; FIG. 2A illustrates aspects of data plane separation in networks to reuse virtual network addresses for multiple hosts, according to at least one embodiment; FIG. 2B illustrates IB management aspects of a system for data plane separation in networks to reuse virtual network addresses and for local port grouping for rails of network links, according to at least one embodiment; FIG. 3 illustrates further aspects of a system for data plane separation and for local port grouping for rails of network links in IB networks using a subnet manager (SM), according to at least one embodiment; FIG. 4 illustrates still further aspects of a system for data plane separation and for local port grouping using an SM having a mapping of different virtual network addresses and a relationship between the different virtual network addresses, according to at least one embodiment; FIG. 5A illustrates a process flow for data plane separation in networks to reuse virtual network addresses, according to at least one embodiment; FIG. 5B illustrates a process flow for data plane separation and for local port grouping in networks to enable exclusive communication, according to at least one embodiment; FIG. 6A illustrates process flow to support data plane separation in networks to reuse virtual network addresses, according to at least one embodiment; FIG. 6B illustrates another process flow for data plane separation and for local port grouping in networks to enable exclusive communication, according to at least one embodiment; FIG. 7A illustrates yet another process flow for data plane separation in networks to reuse virtual network addresses, according to at least one embodiment; and FIG. 7B illustrates yet another process flow for data plane separation and for local port grouping in networks to enable exclusive communication, according to at least one embodiment. DETAILED DESCRIPTION In at least one embodiment, FIG. 1 illustrates a system 100 that is subject to data plane separation in networks to reuse virtual network addresses. Further, local port grouping for rails of network links in IB networks may be supported using the reused virtual network addresses, as detailed herein. A local port is used interchangeably with a physical port herein. The system 100 enables a data plane separation in networks to reuse virtual network addresses (such as local identifiers or LIDs) for different physical ports of a host machine. The different physical ports are associated with different switches of different data planes so that traffic of the host machine can be communicated concurrently and separately using the different physical ports that is addressed using a same virtual network address, for instance; and so that the same virtual network address can be reused in the different switches. In at least one embodiment, a system 100 also enables virtual data plane separation in networks to reuse virtual network addresses. For example, instead of data plane separation that may include physical separation (such as, using different switches), logical or virtual data plane separation may be enabled, in part by rail optimization, in which a switch can be divided into “virtual switches” with a logical separation that is implemented by forwarding logic and without a need for physical separation. Therefore, as used herein, data plane separation may be physical o