摘要:
The invention replicates a packet requiring high availability and transmits it from two or more ports of a switch, for example a wiring closet Layer 2 switch. The parent packet carries a unique sequence number. The copies of the packet each carry the parent packet's unique sequence number. Each copy of the packet then travels on separate pathways through routers (Layer 3 network devices). The pathways are maintained separate by assigning high costs in a LSP routing sense to links connecting the two paths, and by assigning low costs to links along the desired paths. The two identical packets converge on the destination station. The destination station accepts the first packet with a particular sequence number, and discards any later packets with the same sequence number. In the event that a link in one path has a catastrophic failure, then the packet travelling along the other path reaches the destination station and service remains operative without interruption. The lost path is then recomputed by the a router (if any) still receiving the lost packet. A new non-converging path may be selected if any are available, depending upon the topology. After the disruption is repaired, the original dual pathways may once again be established.
摘要:
A network traffic shaper provides high-speed, multi-level shaping. The traffic shaper is in communicating relationship with a forwarding engine, and includes a queue controller having a plurality of queues for storing messages, a scheduler for computing release times, at least one time-searchable memory and a corresponding memory controller. Each queue is preferably associated with a corresponding traffic specifier, and a release time is computed for each queue and stored in the time-searchable memory. When a stored release time expires, the message at the head of the corresponding queue is retrieved and is either moved into a different queue or forwarded by the network device. By moving messages through two or more queues, each having its own release time computed in response to a different traffic specifier, the traffic shaper can perform multi-level shaping on network messages.
摘要:
A method for searching network messages for pre-defined regular expressions is disclosed. A plurality of pre-defined regular expressions are stored in a content-addressable memory (CAM). A network message or selected portion thereof is inputted to the CAM for comparison with all of the regular expressions stored therein, the comparison with all CAM entries being done at the same time. An output is returned from the CAM. In response to the output from the CAM, identifying an action to be applied to the given network message or portion thereof that corresponds to a CAM entry matching the inputted network message or selected portion thereof.
摘要:
A network traffic shaper includes a traffic shaper table for storing traffic specifiers, such as permissible data transmission rates, an arithmetic logic unit (ALU), and a high-speed forwarding trigger mechanism having at least one time-searchable data structure or queue and a retrieve time generator that substantially tracks, but never exceeds, a system time. As network messages are received, they are stored at a message buffer and certain message parameters, including message length and a corresponding traffic specifier, are provided to the traffic shaper. The traffic shaper determines when the message may be sent in accordance with the associated traffic specifier and stores this transmission start time along with the message's buffer location in the time-searchable queue of the forwarding trigger. The forwarding trigger continuously examines the transmission start times for previously stored messages. When the transmission start time of a given message matches the retrieve time generator, the forwarding trigger mechanism signals that the message may be sent.
摘要:
A programmable pattern matching engine efficiently parses the contents of network messages for regular expressions and executes pre-defined actions or treatments on those messages that match the regular expressions. The pattern matching engine is preferably a logic circuit designed to perform its pattern matching and execution functions at high speed, e.g., at multi-gigabit per second rates. It includes, among other things, a message buffer for storing the message being evaluated, a decoder circuit for decoding and executing corresponding actions or treatments, and one or more content-addressable memories (CAMs) that are programmed to store the regular expressions used to search the message. The CAM may be associated with a second memory device, such as a random access memory (RAM), as necessary, that is programmed to contain the respective actions or treatments to be applied to messages matching the corresponding CAM entries. The RAM provides its output to the decoder circuit, which, in response, decodes and executes the specified action or treatment.
摘要:
A network traffic shaper for shapping transmission of network messages includes a system time generator for generating a system time, an arithmetic logic unit (ALU) for computing a transmission start time for each network message in response to the system time, and a retrieve time generator adapted to increment a retrieve time at a rate faster than the system time. As network messages are received, they are stored in a queue along with an associated transmission start time for each message. A forwarding trigger transmits a store network messages when its associated transmission start time matches the retrieve time. Alternately, a second transmission start time representing an excess bandwidth transmission start time may be computed for each network message. If excess bandwidth is detected, a message may be transmitted when its second transmission start time matches the retrieve time.
摘要:
An IP packet reassembly engine provides high-speed and efficient reassembly of IP fragments received at an intermediate station in a computer network. The IP packet reassembly engine comprises a main controller logic circuit configured to “speed-up” re-assembly of original packets from IP fragments stored in a frame buffer at multi-gigabit per second rates. To that end, the reassembly engine further includes a content addressable memory having a plurality of entries for maintaining status information for each received fragment and for each original packet being reassembled from the fragments.
摘要:
The present invention provides methods and devices for implementing a Low Latency Ethernet (“LLE”) solution, also referred to herein as a Data Center Ethernet (“DCE”) solution, which simplifies the connectivity of data centers and provides a high bandwidth, low latency network for carrying Ethernet and storage traffic. Some aspects of the invention involve transforming FC frames into a format suitable for transport on an Ethernet. Some preferred implementations of the invention implement multiple virtual lanes (“VLs”) in a single physical connection of a data center or similar network. Some VLs are “drop” VLs, with Ethernet-like behavior, and others are “no-drop” lanes with FC-like behavior. Some preferred implementations of the invention provide guaranteed bandwidth based on credits and VL. Active buffer management allows for both high reliability and low latency while using small frame buffers. Preferably, the rules for active buffer management are different for drop and no drop VLs.
摘要:
A Data Center Ethernet (“DCE”) network and related methods and device are provided. A DCE network simplifies the connectivity of data centers and provides a high bandwidth, low latency network for carrying Ethernet, storage and other traffic. A Fiber Channel (“FC”) frame, including FC addressing information, is encapsulated in an Ethernet frame for transmission on a Data Center Ethernet (“DCE”) network. The Ethernet address fields may indicate that the frame includes an FC frame, e.g., by the use of a predetermined Organization Unique Identifier (“OUI”) code in the D_MAC field, but also include Ethernet addressing information. Accordingly, the encapsulated frames can be forwarded properly by switches in the DCE network whether or not these switches are configured for operation according to the FC protocol. Accordingly, only a subset of Ethernet switches in the DCE network needs to be FC-enabled. Only switches so configured will require an FC Domain_ID.
摘要:
The present invention provides methods and devices for implementing a Low Latency Ethernet (“LLE”) solution, also referred to herein as a Data Center Ethernet (“DCE”) solution, which simplifies the connectivity of data centers and provides a high bandwidth, low latency network for carrying Ethernet and storage traffic. Some aspects of the invention involve transforming FC frames into a format suitable for transport on an Ethernet. Some preferred implementations of the invention implement multiple virtual lanes (“VLs”) in a single physical connection of a data center or similar network. Some VLs are “drop” VLs, with Ethernet-like behavior, and others are “no-drop” lanes with FC-like behavior. Some preferred implementations of the invention provide guaranteed bandwidth based on credits and VL. Active buffer management allows for both high reliability and low latency while using small frame buffers. Preferably, the rules for active buffer management are different for drop and no drop VLs.