Broadcom Corporation

Martin Lund

Howard A Baumer

Mohan V Kalkunte

Shekhar Ambe

Uri Elzur

Shiri Kadambi

Daniel Talayco

Sandeep Kumar Relan

Allan Christie

Gregory Young

Nicholas Ilyadis

Ariel Hendel

A universal network interface controller (UNIC) is provided for interfacing a host computer to a switch fabric, a packet network, or both. The UNIC includes ingress transmit logic designed to transmit switch fabric data in memory associated with the host computer to a switch fabric. The UNIC further includes egress receive logic designed to receive switch fabric data from the switch fabric to store the received switch fabric data in the host memory associated with the host computer. As an option, the ingress transmit logic may be further designed to transmit packet network data in memory associated with the host computer to a packet network, such as Ethernet, and the egress receive logic may be further designed to receive the packet network data from the packet network and to store the received switch fabric data and the received packet network data in the host memory associated with the host computer.

Universal network interface controller

A method for establishing a virtual channel between network devices is disclosed. In the case of a local network device establishing a virtual channel with a remote network device, a virtual channel request message is sent from the local network device to the remote network device. A virtual channel acknowledgement message and a remote capability list are received and a virtual channel resume message and a local capability list are sent. The virtual channel is then enabled. In the case of a remote network device establishing a virtual channel with a local network device, a virtual channel request message is received from a local network device by a remote network device. A virtual channel acknowledgement message and a remote capability list are sent and a virtual channel resume message and a local capability list are received. The virtual channel is then enabled.

Multiple logical channels for use in network devices

Certain aspects of a method and system for intrusion detection and prevention based on packet type recognition in a network are disclosed. Aspects of one method may include determining a packet type for each of a plurality of received network packets based on at least one of: a header and content of each of the plurality of received network packets. The rate at which the plurality of received network packets are handled at a port in the network switching device may be regulated based on a number of occurrences of the determined packet type of the plurality of received network packets.

Method and system for intrusion detection and prevention based on packet type recognition in a network

Certain aspects of a method and system for adaptive queue and buffer control based on monitoring in a packet network switch are disclosed. Aspects of one method may include pre-allocating packet buffers from a pool of packet buffers in a network switching device based on anticipation of a type of application data being handled by a particular port of the network switching device according to packet classification and tracking. The buffers may be dynamically allocated and deallocated from the pool of packet buffers for a particular port of the network switching device based on changes to the application data.

Method and system for adaptive queue and buffer control based on monitoring in a packet network switch

Certain aspects of a method and system for a power control based on application awareness in a packet network switch are provided. Data communication flow may be monitored in ports in a packet network switch based on packet classification. Ports where data flow is not detected may have at least some functionality disabled to reduce power consumption. In this regard, a power saving mode may be utilized for disabling at least some functionality in a switch port, such as Ethernet ports, for example. A partially disabled port may be fully enabled when monitoring detects active data communication flow in that port. Port functionality may be enabled or disabled sequentially, for example. In some instances, a physical layer portion of the packet network switch may be utilized to adjust power in a port based on the data communication flow.

Method and system for power control based on application awareness in a packet network switch

A method and system of providing physical port security in a digital data network is disclosed. The system keeps bit maps of allowed physical output ports for each physical network connection. The map of allowed ports can be different for different source addresses connected to the device. When digital data, such as an IP packet, is received, the appropriate physical port security bit map is retrieved and a logical AND is done on the physical port bit map generated by the destination information. The resulting bit map is used to determine which physical ports the data is routed to, blocking any requested destinations that are not appropriate destinations based on the port security bit map.

Method and system to provide physical port security in a digital communication system

Certain aspects of a method and system for an adaptive queue and buffer control based on monitoring and active congestion avoidance in a packet network switch are provided. When memory from a memory source is not available for allocation to one or more adaptive queues in a network device, such as a network switch, the TCP window size associated with the one ore more adaptive queues may be reduced. The window size may be reduced to a target size by utilizing one or more reduction steps, wherein the reduction steps may be programmable. When portions of the memory source become available for allocation to the adaptive queues, the window size may be increased. The window size may be increased to a target size by utilizing one or more increase steps, wherein the increase steps may be programmable.

Method and system for adaptive queue and buffer control based on monitoring and active congestion avoidance in a packet network switch

Multiple virtual channels for use in network devices

A method for processing network information in a multi-server platform is disclosed and includes receiving by a blade server manager, capacity utilization information embedded in spare link bandwidth from a plurality of blade servers operably coupled to the blade server manager. Digital information received on a digital communication link may be forwarded to one of the plurality of blade servers. The blade servers may be selected based on the received capacity utilization information. The capacity utilization information may include blade server work load capacity, blade server power saving capacity, blade server CPU percent utilization, and/or data representing blade server interrupt utilization. The blade server manager may be operably coupled to an Ethernet network and/or an external network.

Method and system to provide blade server load balancing using spare link bandwidth

A method for communication is disclosed and may include performing, using one or more processors and/or circuits in a server platform, negotiating a rate at which one or more of a plurality of blade servers in the server platform communicate. Data may be communicated at the negotiated data rate by the one or more of the plurality of blade servers. The data rate may be negotiated in response to one or more packets received by the one or more of the plurality of blade servers. At least a portion of the data may be communicated from a first of the plurality of blade servers to a second of the plurality of blade servers at the negotiated data rate.

System and method for communicating using a multiserver platform

Aspects of the invention provide a method and system for coding information in a communication channel. More particularly, aspects of the invention provide an method and system for synchronous running encryption and/or encoding and corresponding decryption and decoding in a communication channel or link. Aspects of the method may include encoding and/or encrypting a first data using a first or second encoding table and/or a first or second encryption table. The method may indicate which one of the first or second encoding tables or which one of the first or second encryption tables were utilized for encoding and/or encrypting the said first data. The encoded and/or encrypted first data may subsequently be transferred downstream and decoded by synchronous decoder/decryptor using a corresponding decoding and/or decryption table. The corresponding decoding and/or decryption table may be determined based on the indicated first and/or second encoding and/or encrypting tables

Method and system for providing synchronous running encoding and encryption

Aspects of the invention may include receiving a packet from a first of a plurality of blade servers coupled via a common bus. An identifier associated with a second blade server may be determined based on at least a portion of the received packet and at least a portion of the received packet may be routed to at least the second blade server. At least a portion of the received packet may be transferred to at least the second blade server via the common bus. Routing of at least a portion of the received packet may be controlled by a switch blade which may also be coupled to the common bus. In order to route information in the server, an identifier of the first blade server, the second blade server and the switch blade may be determined. Exemplary identifiers may be a MAC address and/or an IP address.

System and method for communicating between servers using a multi-server platform

A blade server with an improved method and apparatus for controlling the capacity utilization of the servers is disclosed. Capacity utilization information is obtained from blade servers mounted on a common backplane. The servers provide information concerning capacity utilization using spare link bandwidth without polling or the use of special messages. The blade manager then allocates network traffic based on a utilization algorithm in order to balance the capacity of the server in an efficient manner.

A system for encoding data in a multilane communication channel may include at least one processor operable to generate, from existing control characters in a character set, expanded control characters utilized for controlling the data in each lane of the multilane communication channel. Each lane of the multilane communication channel may transport the data in a similar direction. The at least one processor is also operable to control at least one of the lanes of the multilane communication channel using at least one of the generated control characters. If a first control character of the existing control characters is a start-of-packet control character, the at least one processor is then operable to select a second control character from any other of the generated expanded control characters, and to indicate a start of a packet using the selected second control character for at least one of the lanes.

Growing community of inventors

Menlo Park, CA, United States of America

Martin Lund