International Business Machines Corporation

Extreme Networks, Inc.

Edward Joel Rovner

Cedell Adam Alexander, Jr

John Kevin Frick

Donald Bruce Grosser, Jr

Matthew B Squire

Olen L Stokes

Deepak Vig

Cedell A Alexander

Charles Allen Carriker

Brian Mitchell Bass

Michael Yip

John T Lloyd

Kenneth Harvey Potter, Jr

Sonia Kiang Rovner

Jim Philip Ervin

Daniel Dean Heimsoth

Scott Nellenbach

Hood L Richardson, Jr

Rosemary Venema Slager

Desikan Saravanan

Robert William Bartoldus

Lance Richardson

Brian W Bailey

J Kevin Frick

Timothy Droz

Brian N Bingham

John Edward Dickey

Arthur Lin

Tash Hepting

Paul Anderson

Justus W Gries

Jon Michael Houghton

Mark Alan Tippens

Jackie Jackson, Jr

The subject matter described herein includes a packet forwarding device that implements next hop scaling. Rather than storing a complete set of next hop bindings at each packet processor, the storage of next hop bindings is distributed among packet processors in the packet forwarding device such that each packet processor stores next hop bindings for the hosts that are directly connected to the packet processor. For hosts that are not directly connected to a packet processor, the packet processor stores relay entries. Because of the distributed storage of next hop bindings, the number of hosts that can be served by a single packet forwarding device is increased over packet forwarding devices where each packet processor stores a complete set of next hop bindings for all connected hosts.

Methods, systems, and computer readable media for next hop scaling

The subject matter described herein includes methods, systems, and computer readable media for next hop scaling with link aggregation. According to one aspect of the subject matter described herein, a system for next hop scaling is provided. The system includes a packet forwarding device including a plurality of packet processors for performing next hop and link aggregation group (LAG) selection operations. Within this plurality of packet processors, ingress packet processors are configured to indicate, for received packets that have a next hop on a different packet processor, that an egress next hop selection operation is needed. Egress packet processors of the plurality of packet processors are configured to perform the egress next hop and member selection operations for the packets for which an egress next hop selection operation is indicated, wherein forwarding of the packets is limited to active LAG group members local to the egress packet processor.

Methods, systems, and computer readable media for next hop scaling with link aggregation

The subject mailer described herein includes methods, systems, and computer readable media for automatically selecting between Internet protocol switching modes on a per-module basis in a packet forwarding device. According to one aspect, a method may include determining capacities of hardware longest prefix matching (LPM) tables located on each input/output (I/O) module in a multi-module IP packet forward device. The number of routes currently stored in a software LPM table may be determined. If the software LPM table can be stored within the hardware LPM table for an I/O module, an LPM mode may be automatically selected for that I/O module. If the contents of software LPM table cannot be stored within the hardware LPM table for a particular I/O module, the I/O module may be automatically transitioned to operate in an Internet protocol forwarding database (IPFDB) mode.

Methods, systems, and computer readable media for automatically selecting between internet protocol switching modes on a per-module basis in a packet forwarding device

An indication of a host route to be added to a forwarding database table as an entry is received. The host route is added to a first hardware table or a second hardware table if a space is available in the second hardware table or in a first storage area of the first hardware table. The first hardware table has both a first storage area and a second storage area. If a space is not available in the second hardware table or the first storage area of the first hardware table, the first storage area of the first hardware table is automatically expanded to include unused space in the second storage area of the first hardware table. The host route is then added to a space in the expanded first storage area of the first hardware table.

Method and system for automatic expansion and contraction of IP host forwarding database

The subject matter described herein includes methods, systems, and computer program products for routing packets at a multi-mode layer 3 packet forwarding device. According to one aspect, the subject matter described herein includes operating a first of at least two modules in a host mode, and operating a second of at least two modules in a longest prefix matching (LPM) mode. Operating a module in a host mode includes populating a host table and an LPM table with entries corresponding to hosts and routing layer 3 packets received by the first module using the host and LPM tables. Operating a module in an LPM mode includes populating a host table with entries corresponding to hosts, populating an LPM table with entries corresponding to variable length Internet protocol (IP) addresses and next hop addresses, and routing layer 3 packets received by the second module using the host and LPM tables.

Methods, systems, and computer program products for routing packets at a multi-mode layer 3 packet forwarding device

Methods, systems, and computer program products for controlling updating of a layer 3 host table based on packet forwarding miss counts are disclosed. According to one method, layer 3 packets are routed using at least one of a layer 3 host table containing entries corresponding to remote hosts and a longest prefix matching table containing prefixes corresponding to remote hosts. For each layer 3 destination address for which a lookup in at least one table fails, a number of packets received within a time period are counted. Remote destination entries in the host table are replaced based on the counts.

Methods, systems, and computer program products for controlling updating of a layer 3 host table based on packet forwarding lookup miss counts

A network switch receives a data packet, recognizes the data packet as belonging to a given stream, and directs the data packet to a stream processing module in the network switch in response to recognizing the stream to which the packet belongs. The stream processing module further recognizes the source of the stream from which the packet was transmitted, and stores the data packet in a buffer for later transmission, as controlled by the stream processing module, to a destination for the stream. In this way, the stream may be replicated and transmitted by the stream processing module any number of times to multiple destinations in an efficient, cost-effective manner.

Method and apparatus providing for delivery of streaming media

A method for forwarding data packets to one of a plurality of servers comprising receiving a data packet from a source, at a data packet forwarding device having a plurality of ports, performing a hashing function using a unique component of the data packet as a seed value for the hash, generating a hash value using the hashing function, looking up a table for an address of one of the plurality of servers using the hashed value as an index to the table, and forwarding the data packet to a server corresponding to the address.

Method and apparatus for server load balancing

Methods and systems for efficiently delivering data to multiple destinations in a computer network are disclosed. Data to be delivered to multiple destinations is stored in memory. The stored data is divided into blocks. Checksums are pre-calculated for each block. Pre-calculated checksums are used to calculate the data checksums for data to be inserted in each packet to be delivered to data destinations.

Methods and systems for efficiently delivering data to a plurality of destinations in a computer network

A method and system for providing enhanced peer redundancy in an ATM emulated local area network (ELAN) served by a primary LAN emulation server (LES), a peer LAN emulation server, a broadcast and unknown server (BUS), and a LAN emulation configuration server (LECS). An enhanced peer redundancy virtual channel connection (VCC) is established between the primary LES/BUS and the peer LES/BUS. The primary LES/BUS and the peer LES/BUS exchange status messages every two seconds indicating the number of assigned LECs. If the primary has fewer LECs than the peer, the primary yields to the peer as the active LES/BUS. If an enhanced redundancy VCC was not established, the backup LES/BUS provides ELAN services while waiting for the primary to call to establish an enhanced redundancy VCC. If an enhanced peer redundancy VCC cannot be established, the primary attempts to establish a redundancy VCC with the peer in which the peer LES/BUS is kept in a dormant mode as long as the redundancy VCC is present. If the redundancy VCC is absent, the peer LES/BUS is activated to service the ELAN.

Method and system for providing peer redundancy to asynchronous transfer mode emulated local-area networks

A LAN interconnect device includes a plurality of Frame Processing Units (FPUs) for coupling each port of the device to a switch fabric. Each one of the Frame Processing Units includes an input section with input logic which prepares LS Headers and appends each one to a block of the frame as the block is forwarded to the switch fabric. The FPU, also, includes an output section with copy logic for copying and assembling frames to be forwarded to devices connected to the port. The copy decision is based upon the LS Header and configuration information in the port.

LAN frame copy decision for LAN switches

A data communication network includes a network client, a network server, and a name caching entity located between the network client and the network server. The network server is coupled to the data communication network by a plurality of network adapters that each have a different network (e.g., Media Access Control (MAC)) address, but share a common logical name. The name caching entity includes a name cache, which stores the common logical name shared by the server's network adapters in association with the network address of one of the network adapters. By reference to the name cache, the name caching entity is able to convert broadcast frames to unicast frames, thereby advantageously reducing broadcast frames in the network. The name caching entity will disassociate the logical name and the network address in the name cache in response to a failure to detect a frame indicating the establishment of a session with the network server within a selected interval of transmission of a frame requesting establishment of a session with the network server. In this manner, the network client can establish a session with the network server through another network adapter in cases in which the network adapter whose network address is stored in the name cache cannot support an additional session.

Method and system for broadcast management in a data communication network that permits namesharing

A method for providing interoperability between multiple versions of LUNI within a mixed Asynchronous Transfer Mode emulated LAN is disclosed. The mixed ATM emulated LAN includes multiple LAN emulated clients (LE clients) having different versions of LAN Emulation User Network Interface (LUNI), such as a first version LUNI and a second version LUNI. The mixed ATM emulated LAN is served by a LAN Emulation Server (LES), a Broadcast and Unknown Server (BUS), and a LAN Emulation Configuration Server (LECS). When an LE Control Frame is being sent by an LE client having a second version LUNI, a determination is made as to whether or not the LE Control Frame will be received by an LE client having a first version LUNI. In response to a determination that the LE Control Frame sent by the LE client having a second version LUNI will be received by an LE client having a first version LUNI, the LE Control Frame is copied, and the copied LE Control Frame is converted to an LE Control Frame with a format that is readable by the LE client having a first version LUNI.

Method and system for providing interoperability between network clients having different versions of local-area network emulation user network interface within an asynchronous transfer mode emulated local-area network

A method of alleviating network congestion in an emulated Local Area Network (ELAN) within an Asynchronous Transfer Mode (ATM) network is implemented. The method mitigates network congestion by distributing the connection times of Local Area Network (LAN) Emulation Client (LEC) to an ELAN over random periods of time. When a LAN Emulation Server (LES) or Broadcast and Unknown Server (BUS) attempts to build point-to-multipoint connections to the LEC, it issues control messages to network switches. If the messages are dropped or rejected the LES or BUS recognizes the network is in a congested state and randomly retries to attempt connection.

LAN emulation service delay of multicast tree generation during congestion

A method for providing redundancy to an Asynchronous Transfer Mode (ATM) emulated local-area network (LAN) is disclosed. The ATM emulated LAN is served by a primary LAN Emulation Server, a Broadcast and Unknown Server (LES/BUS), and a LAN Emulation Configuration Server (LECS). A backup LES/BUS is coupled to the ATM emulated LAN. A logical redundancy virtual channel connection (VCC) is then established between the primary LES/BUS and the backup LES/BUS. When the redundancy VCC is present between the primary LES/BUS and the backup LES/BUS, the backup LES/BUS is placed in a dormant mode. However, when the redundancy VCC is absence between the primary LES/BUS and the backup LES/BUS, the backup LES/BUS is activated to serve any request generated by the ATM emulated LAN.

Growing community of inventors

Chapel Hill, NC, United States of America

Edward Joel Rovner