Switch fabric system and data switching method

US10505867B2 · US · B2

Patent metadata
FieldValue
Publication numberUS-10505867-B2
Application numberUS-201715851283-A
CountryUS
Kind codeB2
Filing dateDec 21, 2017
Priority dateJun 25, 2015
Publication dateDec 10, 2019
Grant dateDec 10, 2019

How to read this patent

A practical reading order for non-experts. Skip the full description unless you need deep technical detail.

  1. Title

    What the patent document calls the invention.

  2. Abstract

    A short plain-language summary of the technical disclosure.

  3. Assignees and inventors

    Who owns or filed the patent and who is credited as inventor.

  4. Key dates

    Filing, priority, publication, and grant dates set the timeline.

  5. First independent claim

    The legal scope of protection — read this for what is actually claimed.

  6. CPC / IPC classifications

    Technology tags used to group this patent with similar filings.

  7. Citations and related patents

    Prior art links and similar publications in this corpus.

Abstract

Official abstract text for this publication.

The present disclosure provides a switch fabric system, the system including M first crossbar units (CUs) and N second CUs, where each first CU includes L first input ports, a first arbiter, a first crossbar, and N first output ports. Each second CU includes M second input ports, a second arbiter, a second crossbar, and one second output port. M×N first output ports of the M first CUs are respectively coupled to N×M second input ports of the N second CUs, where N first output ports of each first CU are respectively coupled to and in a one-to-one correspondence with one second input port of each second CU in the N second CUs. In the example system, N equals M×L, and M, N, and L are all positive integers.

First claim

Opening claim text (preview).

What is claimed is: 1. A switch fabric system, comprising: M first crossbar units (CUs) and N second CUs, wherein each first CU of the M first CUs comprises: L first input ports, a first arbiter, a first crossbar, and N first output ports, and wherein each second CU of the N second CUs comprises: M second input ports, a second arbiter, a second crossbar, and one second output port, and wherein M×N first output ports of the M first CUs are respectively coupled to N×M second input ports of the N second CUs, wherein N first output ports of each first CU are respectively coupled to and in a one-to-one correspondence with one second input port of each second CU in the N second CUs, wherein N=M×L, and M, N, and L are all positive integers; wherein any first input port in each first CU is configured to receive and cache data, and generate a first scheduling request according to the received data, wherein the data carries a destination port, the destination port is a second output port of any second CU in the N second CUs, and the first scheduling request is used to request to send the data to the destination port by using the N first output ports; the first arbiter is configured to perform scheduling and arbitration on the first scheduling request to determine a first target output port that matches the destination port and generate a first scheduling grant, wherein the first target output port is one first output port coupled to the any second CU in N first output ports of each first CU, and the first scheduling grant is used to instruct the any first input port to send the data to the first target output port; the any first input port is further configured to schedule the data to the first crossbar according to the first scheduling grant; and the first crossbar is configured to switch the data to the first target output port under configuration of the first arbiter; and a second target input port in the any second CU is configured to receive and cache the data, and generate a second scheduling request, wherein the second target input port is one second input port in the any second CU and coupled to the first target output port, and the second scheduling request is used to request to use the second output port of the any second CU; the second arbiter is configured to perform arbitration on the second scheduling request to generate a second scheduling grant, wherein the second scheduling grant is used to instruct the second target input port to send the data to the second crossbar; the second target input port is further configured to schedule the data to the second crossbar according to the second scheduling grant; and the second crossbar is configured to switch the data to the second output port of the any second CU under configuration of the second arbiter. 2. The switch fabric system according to claim 1 , wherein the first arbiter is configured to perform scheduling and arbitration on the first scheduling request according to a preset mapping relationship to determine, from N first output ports of each first CU, the first target output port that matches the destination port, wherein the preset mapping relationship comprises a one-to-one correspondence between N first output ports of each first CU and N second output ports of the N second CUs. 3. The switch fabric system according to claim 1 , wherein the second arbiter is configured to perform arbitration on the second scheduling request according to a preset scheduling algorithm to generate the second scheduling grant, wherein the preset scheduling algorithm comprises at least one of round-robin (RR) scheduling, oldest cell first (OCF) scheduling, longest queue first (LQF) scheduling, longest port first (LPF) scheduling, and weighted scheduling. 4. The switch fabric system according to claim 1 , wherein each first CU further comprises L first caches, wherein the L first caches are respectively coupled to and in a one-to-one correspondence with the L first input ports of each first CU, and wherein the L first caches are configured to respectively cache data received by the L first input ports of each first CU. 5. The switch fabric system according to claim 1 , wherein each second CU further comprises M second caches, wherein the M second caches are respectively coupled to and in a one-to-one correspondence with the M second input ports of each second CU, and wherein the second cache is configured to cache data received by the second input port. 6. The switch fabric system according to claim 1 , wherein the first crossbar comprises N first multiplexers, wherein the first multiplexer comprises L first ingress ports and one first egress port, wherein the L first ingress ports are respectively coupled to the L first input ports one by one, wherein the first egress port is coupled to one first output port in the N first output ports, and wherein the N first multiplexers are configured to implement channel gating between the any first input port and the first target output port under configuration of the first arbiter to switch the data to the first target output port. 7. The switch fabric system according to claim 1 , wherein the second crossbar comprises a second multiplexer, wherein the second multiplexer comprises M second ingress ports and one second egress port, wherein each of the M second ingress ports is respectively coupled to one second input port of the M second input ports, wherein the second egress port is coupled to the second output port, and wherein the second multiplexer is configured to implement channel gating between the second target input port and the second output port of the any second CU under configuration of the second arbiter to switch the data to the second output port of the any second CU. 8. A data switching method, wherein the method is applicable to a switch fabric system, and the switch fabric system comprises: M first crossbar units (CUs) and N second CUs, wherein each first CU of the M first CUs comprises: L first input ports, a first arbiter, a first crossbar, and N first output ports, and wherein each second CU of the N second CUs comprises: M second input ports, a second arbiter, a second crossbar, and one second output port, and wherein M×N first output ports of the M first CUs are respectively coupled to N×M second input ports of the N second CUs, wherein N first output ports of each first CU are respectively, coupled to and in a one-to-one correspondence with one second input port of each second CU in the N second CUs, and wherein N=M×L, and M, N, and L are all positive integers; wherein the method comprises: receiving and caching, by any first input port in each first CU, data and generating a first scheduling request according to the received data, wherein the data carries a destination port, the destination port is a second output port of any second CU in the N second CUs, and the first scheduling request is used to request to send the data to the destination port by using the N first output ports; performing, by the first arbiter, scheduling and arbitration on the first scheduling request to determine a first target output port that matches the destination port and generate a first scheduling grant, wherein the first target output port is one first output port coupled to the any second CU in N first output ports of each first CU, and the first scheduling grant is used to instruct the any first input port to send the data to the first target output port; scheduling, by the any first input port, the data to the first crossbar according to the first scheduling grant; switching, by the first crossbar, the data to the first target output port under configuration of the first arbiter; receiving and caching, by a second target input port in the any second CU, the data and generatin

Assignees

Inventors

Classifications

  • based on packet size, e.g. shortest packet first · CPC title

  • Input queuing · CPC title

  • H04L49/101Primary

    using crossbar or matrix · CPC title

  • Fixed service order, e.g. Round Robin · CPC title

  • H04L49/10Primary

    characterised by the switching fabric construction · CPC title

Patent family

Related publications grouped by family.

External sources

Frequently asked questions

Answers are generated from the same data shown on this page.

What does patent US10505867B2 cover?
The present disclosure provides a switch fabric system, the system including M first crossbar units (CUs) and N second CUs, where each first CU includes L first input ports, a first arbiter, a first crossbar, and N first output ports. Each second CU includes M second input ports, a second arbiter, a second crossbar, and one second output port. M×N first output ports of the M first CUs are respe…
Who is the assignee on this patent?
Huawei Tech Co Ltd
What technology area does this patent fall under?
Primary CPC classification H04L49/101. Mapped technology areas include Electricity.
When was this patent published?
Publication date Tue Dec 10 2019 00:00:00 GMT+0000 (Coordinated Universal Time) (B2). Legal status and post-grant events are not shown on this page.
What related patents are in patentsdb?
We list 1 related publication on this page (citations in our corpus or others sharing the same primary CPC).