Method and apparatus to distribute data center network traffic

US9331946B2 · US · B2

Patent metadata
FieldValue
Publication numberUS-9331946-B2
Application numberUS-201313736243-A
CountryUS
Kind codeB2
Filing dateJan 8, 2013
Priority dateJan 8, 2013
Publication dateMay 3, 2016
Grant dateMay 3, 2016

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.

Exemplary embodiments provide a technique to improve the system availability of the systems that have multiple links and are connected to a network fabric. In one embodiment, a switch comprises: a memory storing a first logical group which has a first plurality of IP addresses of a first plurality of ports and is assigned to a first ID (identifier), and a second logical group which has a second plurality of IP addresses of a second plurality of ports and is assigned to a second ID; and a controller controlling to cause a logical path of the first logical group to use a first physical path which is different from a second physical path to be used by a logical path of the second logical group, based on the first ID and the second ID.

First claim

Opening claim text (preview).

What is claimed is: 1. A method of controlling network traffic using a port switch that includes a memory and a controller, the method comprising: storing, in the memory: distributed route information comprising at least two system identifiers (IDs) and at least two distributed route IDs, wherein a route ID identifies a specific physical path from multiple physical paths of a network fabric; a first logical group which has a first plurality of IP addresses of a first plurality of ports and is assigned to a first system ID, wherein the first logical group is assigned to the first system ID by correlating the IP addresses of the first plurality of ports in the first logical group with the first system ID, the first plurality of ports are assigned to a first route ID, the first system ID corresponds to the first route ID, and the first system ID identifies a first fabric switch; and a second logical group which has a second plurality of IP addresses of a second plurality of ports and is assigned to a second system ID, wherein the second logical group is assigned to the second system ID by correlating the IP addresses of the second plurality of ports in the second logical group with the second system ID, the second plurality of ports are assigned to a second route ID, the second system ID corresponds to the second route ID, and the second system ID identifies a second fabric switch; and receiving a plurality of packets, and for each received packet: reading a source IP address of the received packet, and determining whether the source IP address matches an IP address of the first plurality of ports or second plurality of ports; upon determining that the source IP address matches an IP address of the first plurality of ports or the second plurality of ports, determining whether the source IP address is assigned to the first system ID or the second system ID; upon determining that the source IP address is assigned to the first system ID, determining an outgoing port for the packet based on the first physical path and causing the first logical group to use the first physical path which is different from a second physical path; and upon determining that the source IP address is assigned to the second system ID, determining an outgoing port for the packet based on the second physical path and causing the second logical group to use the second physical path; upon determining that the source IP address does not match an IP address of the first plurality of ports and the second plurality of ports, reading a destination IP address of the received packet, and determining whether the destination IP address matches an IP address of the first plurality of ports or the second plurality of ports; upon determining that the destination IP address matches an IP address of the first plurality of ports or the second plurality of ports, determining whether the destination IP address is assigned to the first system ID or the second system ID; upon determining that the destination IP address is assigned to the first system ID, determining an outgoing port for the packet based on the first physical path and causing the first logical group to use the first physical path; and upon determining that the destination IP address is assigned to the second system ID, determining an outgoing port for the packet based on the second physical path and causing the second logical group to use the second physical path; and upon determining that the destination IP address matches neither an IP address of the first plurality of ports nor the second plurality of ports, calculating a hash value of the source IP address and the destination IP address of the received packet, and selecting an outgoing port from the distributed route information according to an order of the hash value based on a correlation between order of hash values and order of paths through a network fabric. 2. The method according to claim 1 , wherein correspondence between the first system ID and the first route ID indicates first paths associated with the first route ID and using the first fabric switch, and correspondence between the second system ID and the second route ID indicates second paths associated with the second route ID and using the second fabric switch. 3. The method according to claim 1 , wherein each distributed route ID directly corresponds to an outgoing port. 4. The method according to claim 1 , wherein the first logical group is assigned to the first system ID by correlating VLAN (Virtual Logical Area Network) IDs of the first plurality of ports in the first logical group with the first system ID; and wherein the second logical group is assigned to the second system ID by correlating VLAN IDs of the second plurality of ports in the second logical group with the second system ID. 5. The method according to claim 4 , further comprising, for each packet received: reading a VLAN ID of the received packet; if the VLAN ID of the received packet matches a VLAN ID of the first plurality of ports or the second plurality of ports, determining whether the VLAN ID is assigned to the first system ID or the second system ID; if the VLAN ID is assigned to the first system ID, determining an outgoing port for the packet based on the first physical path; and if the VLAN ID is assigned to the second system ID, determining an outgoing port for the packet based on the second physical path; and if the VLAN ID does not match a VLAN ID of the first plurality of ports and the second plurality of ports, calculating a hash value of the VLAN ID of the packet and select an outgoing port from the distributed route information according to an order of the hash value based on a correlation between order of hash values and order of paths through the network fabric. 6. The method according to claim 1 , wherein the first logical group is assigned to the first system ID by correlating VNIs (VXLAN (Virtual eXtensible LAN) Network Identifiers) of the first plurality of ports in the first logical group with the first system ID; wherein the second logical group is assigned to the second system ID by correlating VNIs of the second plurality of ports in the second logical group with the second system ID; and wherein the method further comprises, for each packet received: reading a VNI of the received packet; if the VNI of the received packet matches a VNI of the first plurality of ports or the second plurality of ports, determining whether the VNI is assigned to the first system ID or the second system ID; if the VNI is assigned to the first system ID, determining an outgoing port for the packet based on the first physical path; and if the VNI is assigned to the second system ID, determining an outgoing port for the packet based on the second physical path; and if the VNI does not match a VNI of the first plurality of ports and the second plurality of ports, calculating a hash value of the VNI of the packet and select an outgoing port from the distributed route information according to an order of the hash value based on a correlation between order of hash values and order of paths through the network fabric. 7. A port switch comprising: a memory storing distributed route information comprising at least two system identifiers (IDs) and at least two distributed route IDs, wherein a route ID identifies a specific physical path from multiple physical paths of a network fabric; a first logical group which has a first plurality of Internet Protocol (IP) addresses of a first plurality of ports and is assigned to a first system ID, wherein the first logical group is assigned to the first system ID by correlating the IP addresses of the first plurality of ports in the first logical group with the first system ID, the first plurality of ports are assign

Assignees

Inventors

Classifications

  • Link aggregation, e.g. trunking · CPC title

  • Electricity · mapped topic

  • H04L47/125Primary

    by balancing the load, e.g. traffic engineering · CPC title

  • Cross-Sectional Technologies · mapped topic

  • across network layers, e.g. resolution of network layer into physical layer addresses or address resolution protocol [ARP] · 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 US9331946B2 cover?
Exemplary embodiments provide a technique to improve the system availability of the systems that have multiple links and are connected to a network fabric. In one embodiment, a switch comprises: a memory storing a first logical group which has a first plurality of IP addresses of a first plurality of ports and is assigned to a first ID (identifier), and a second logical group which has a second…
Who is the assignee on this patent?
Okita Hideki, Hitachi Ltd
What technology area does this patent fall under?
Primary CPC classification H04L47/125. Mapped technology areas include Electricity.
When was this patent published?
Publication date Tue May 03 2016 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 8 related publications on this page (citations in our corpus or others sharing the same primary CPC).