Cisco Technology, Inc.

Qualcomm Incorporated

Ubicom Gmbh

Ankur Goyal

Christian Sasso

Ronak K Desai

Siddharth Kasat

Tejas Kokje

Hariharan Balasubramanian

Shivaranjani Sankara Krishnan

Manuwela Kanade

Samar Sharma

Yibin Yang

Amitkumar V Patel

David J Hudson

Keith J Morris

Anand Girish Parthasarathy

Santosh Rajagopalan

Kishan Kumar Kotari Uppunda

Prashanth Pappu

Vithal Shirodkar

Jaimin Pankajkumar Patel

Kishan Kumar Kotari Uppanda

Upon receiving a first message, from the second network device, indicating that the second network device is incompatible with one or more virtual local area network (VLAN) pruning techniques, a timer on a first network device is configured to expire after a predetermined period of time. The first network device is configured to maintain a subscription for the second network device to one or more VLANs until the timer expires. The second network device is configured to transmit the first message in response to detecting an upcoming control plane outage at the second device. Prior to the timer expiring, embodiments transmit data assigned to the one or more VLANs to the second network device. Embodiments also periodically receive update messages, from the second network device, identifying one or more VLANs to which the second network device wants to subscribe.

Hitless pruning protocol upgrade on single supervisor network devices

In some embodiments, a virtual device context (vDC) domain may be advertised to other network devices. If at least a partition of each device is determined to belong to the same vDC domain, the network interface communicating with the at least one device may be activated.

Auto management of a virtual device context enabled network infrastructure

Certain embodiments of the present disclosure generally relate to allocating a sub-area of Fiber Channel addresses (FCIDs) to a device. A range of addresses may be assigned to the device using a mask address, where the most significant bits represent a mask and the least significant bits represent a sub-range of FCIDs available to be assigned to the device. Therefore, routing information may be stored efficiently in a Ternary Content Addressable Memory (TCAM) by storing a single entry in the TCAM for each sub-area of FCIDs allocated to a device, instead of storing an entry for each FCID. The single entry may indicate the mask address and the width of the mask.

Sub-area FCID allocation scheme

Techniques for configuring a virtual network device are provided. Embodiments identify a network device profile specifying a virtual link teaming domain. Upon receiving a network message advertising the virtual link teaming domain from a peer network device of the two or more network devices, on a plurality of links of the first network device, the plurality of links are teamed, based on the virtual link teaming domain, to create a first link bundle. Embodiments advertise a virtual link teaming identifier to a neighboring network device of the first network device. Here, the neighboring network device is configured to use the virtual link teaming identifier to bundle links between the neighboring network device and the virtual network device.

VPC auto configuration

In Service Software Upgrade (ISSU) permits administrators to upgrade the control plane software of a network device without any disruption to the traffic passing through the data plane of the network device. However, because the control plane is unavailable, the network device is unable to transmit periodic messages to re-subscribe to specific VLANs. This may result in a service outage in end devices that rely on the network device to receive data assigned to those VLANs. To prevent the network device from being unsubscribed from the VLANs while the control plane is unavailable, the network device may transmit a control plane message before starting ISSU. Once a neighboring network device receives the message, the neighboring device delays unsubscribing the network device until the control plane is again available. Thus, the network device may perform ISSU without the possibility of data traffic not reaching end devices.

Techniques for providing an extended layer 2 network. Embodiments receive, at a first network device in a first layer 2 network within the extended layer 2 network, a data packet specifying a destination Media Access Control (MAC) address. Upon determining that the destination MAC address is a site virtual MAC (S-VMAC) address, the data packet is forwarded to a second network device in a second layer 2 network within the extended layer 2 network and associated with the S-VMAC address, where the second network device is configured to determine a second destination MAC address for the data packet based on an Internet Protocol (IP) address specified in the data packet, and is further configured to forward the data packet to a destination device associated with the determined second destination MAC address.

OTV scaling: site virtual MAC address

According to one aspect, a Fiber Channel (FC) switch that is currently serving as a Principal Switch among a plurality of FC switches in a FC Fabric sends an Exchange Fabric Parameters (EFP) frame having an empty domain identifier list to neighboring FC switches during a first portion a Principal Switch Selection phase. During a second portion of the Principal Switch Selection phase that is after the first portion, the first Fiber Channel switch sends to neighboring FC switches an EFP frame having a non-empty domain identifier list to indicate to the neighboring FC switches that the first FC switch is to remain serving as the Principal Switch in the FC Fabric.

Persistent principal switch for fibre channel fabric

A Fiber Channel (FC) switch and related methods are provided for merging the FC switch with an existing FC network. During a merge process of a stand-alone switch with an FC network, a first Exchange Fabric Parameters (EFP) frame is sent from the stand-alone switch to an FC switch in the existing FC network. The first EFP frame has an empty domain identifier (ID) list. The stand-alone switch receives a second EFP frame from the FC switch. The second EFP frame contains a domain ID list of domain IDs already in use in the FC network. The stand-alone switch compares the domain ID list in the second EFP frame with a stored list of one or more domain IDs associated with the stand-alone switch to determine if there is a domain ID overlap or conflict. If no conflict is detected, the stand-alone switch merges with the existing network without initiating a Build Fabric phase or a Principal Switch Selection phase that could cause disruption of traffic in the network.

Merging a stand-alone switch to a fibre channel network

A Fiber Channel switch is provided for a relative addressing scheme for domain values to avoid losing portions of a Fiber Channel identifier. At a Fiber Channel switch, data is stored that assigns a relative domain field value to a run-time domain. The relative domain field value used for a Fiber Channel identifier is determined for devices connected to the switch based on the run-time domain of the devices and the virtual storage area network in which the devices are active. A Fiber Channel identifier is stored in a memory of the switch that comprises the relative domain field value determined for the device together with area and port values for the device. The relative domain value is either a reserved domain field value for an existing virtual storage area network or any of the regular domain field values for a new virtual storage area network.

Domain-independent persistent fibre channel identifiers

In one embodiment, a method includes establishing a link between two N_Port Identifier Virtualization (NPIV) switches, the link having a high cost assigned thereto. The NPIV switches are in communication with a plurality of hosts through an N_Port Virtualization (NPV) device. The method further includes receiving at a first of the NPIV switches, an indication of a failure at a second of the NPIV switches, receiving data at the first NPIV switch, the data destined for one of the hosts associated with a domain of the second NPIV switch, and forwarding the data to the NPV device for delivery to the host, wherein a Fibre Channel Identifier (FCID) of the host is the same before and after the failure at the second NPIV switch. An apparatus is also disclosed.

[object Object]

A local area network includes computers and peripherals networked in a high-speed LAN with access to a WAN through a slower connection via a broadband modem. A LAN gateway device manages data traffic between the local computers and peripherals and between the LAN and the WAN. The LAN gateway device provides multiple features, such as wired or wireless links, security, firewall, NAT, DCHP, traffic management, and the like. Traffic management features include an automatic quality of service priority classification scheme. A quality of service module automatically assigns priorities to the data streams based on analysis of the data packets. A configuration access list can be provided with pre-configured priorities for some streams. Initially, all streams are given highest priority and subsequently the priority is automatically adapted to the results of the packet analysis. Traffic shaping techniques control the LAN gateway upstream output and enable IP fragmentation of TCP packets according to measured upstream channel conditions.

Automatic adaptive network traffic prioritization and shaping

Methods and apparatus for providing in-order delivery in Fibre Channel (FC) fabric are disclosed. A topological change between a first switch and a second switch is evaluated to determine whether the topological change may result in out-of-order delivery. If it is determined that the topological change may result in out-of-order delivery, a flush frame is sent to the second switch and stop-on-mark is performed on all interfaces of affected links. Upon receiving the flush frame, the second switch drains all virtual output queues (VOQs) and sends an acknowledgement frame to the first switch. The first switch resumes in-order-delivery in an affected link after receiving the acknowledgement frame or expiration of a lifetime time of a frame.

Less loss in-order delivery protocol for fibre connection architecture

In one embodiment, the convergence node switches of a destination node switch in a network having multiple equal paths between a source switch and destination switch are identified. When a new equal cost path is added to the network, packets are flushed up to the convergence node switch closest to the source switch.

Reducing packet drops in networks guaranteeing in-order delivery

Systems, methods, and other embodiments associated with computing path costs based on link cost as a primary metric and hop cost as a secondary metric are presented. The secondary metric facilitates distinguishing between what would be equal cost multi-paths if only link cost was employed. One example method includes computing a modified link cost for a plurality of paths between switches. The example method may also include determining a lowest modified link cost associated with a path between the switches. The example method may also include providing a signal identifying the lowest modified link cost.

Growing community of inventors

Sunnyvale, CA, United States of America

Ankur Goyal