Abstract:
An apparatus, computer program and method for packet buffer management in an IP network system. The apparatus includes at least one link queue buffer, a shared buffer, a buffer state detector, and a buffer manager. The at least one link queue buffer is allocated a buffer of a packet stored in a memory. The shared buffer is excessively allocated when exceeding a minimum buffer threshold value. The buffer state detector determines whether a buffer value stored in a link queue buffer of a corresponding link exceeds a preset minimum buffer threshold value. The buffer manager sets the shared buffer to be included in the link queue buffer if the stored buffer value exceeds the preset minimum buffer threshold value.
Abstract:
An input buffer device and control method of the input buffer device. The input buffer device includes a virtual output queuing (VOQ) buffering section which has a plurality of VOQ buffers. The input buffer device stores data which is input to an input port to a VOQ buffer corresponding to an intended output port of the data among the plurality of VOQ buffers. A shared buffering section is provided which stores the data when a VOQ buffer corresponding to the intended output port of the data is full of data. The stored data is forwarded to the VOQ buffer when the VOQ buffer is empty. Accordingly, the input buffer device can more efficiently process the data by use of the fixed-length FIFO buffers and the shared buffer.
Abstract:
A switch for broadcasting packets is provided including a plurality of input ports, a switch fabric coupled to the input ports, a plurality of output ports coupled to the suited fabrics and a multicast interconnect coupled between the input ports and output ports for routing multicast packets directly between the input and output ports. The multicast interconnect may include a multicast queue, an arbiter coupled between the multicast queue and the input ports and a plurality of broadcast buffers coupled to the multicast queue, each broadcast buffer coupled to a corresponding output port. The multicast interconnect may include a plurality of egress arbiters coupled between corresponding broadcast buffers and output ports.
Abstract:
A switch queue predictive protocol (SQPP) includes a packet switching system including: a switch fabric having a cross-point switch, and a plurality of line cards, each coupled to the switch fabric. A cross-point buffer is located at each cross-point of the cross-point switch. The switch fabric also includes a plurality of actual available queue space tables (AAQSTs), each identifying the actual queue space available in a row of the cross-point buffers. Each of the line cards includes an input buffer, an output buffer, and a predicted available queue space table (PAQST) identifying predicted queue space available in a corresponding row of the cross-point buffers. Packet information is transmitted from a source line card to the switch fabric only if available queue space is predicted by the corresponding PAQST. The switch fabric uses the AAQST to update the PAQST after packet information is transmitted to a destination line card.
Abstract:
A network packet may be forwarded by reading a table containing a plurality of flags to determine which of the plurality of flags is set or cleared and performing an operation on the packet to decapsulate or encapsulate the packet in accordance with values of the flags.
Abstract:
A common buffer memory control apparatus controls a common buffer memory which is used to store message data items each of which is divided into a plurality of cells based on an asynchronous transfer mode. The common buffer memory control apparatus includes a free block management table for managing whether each of blocks into which the common buffer memory divided is free or used, a block selecting unit for selecting a block of the common buffer memory which is free with reference to the free block management table, and a cell writing control unit for controlling a write operation for cells of a single message data item so that the respective cells of the single message data item are written in the block, selected by the block selecting means, of the common buffer memory.
Abstract:
The present invention relates to a method and system for supporting in a router a plurality of data flows using a ternary content addressable memory (TCAM) in which the number of accesses to write to the TCAM is optimized to improve efficiency of updating and subsequent look up. To accommodate the plurality of data flows, the TCAM is partitioned into at least two partitions in which a first portion includes indices having a higher priority and a second portion includes indices having a lower priority. For example, multiple protocol label switching (MPLS) flows and IP-Virtual Private Network (VPN) can be added to the first partition and policy based routing flows can be added to the second partition. During subsequent TCAM look-up of a prefix of an incoming packet the MPLS or IP-VPN flow will subsume any matching policy based routing flow, such as flows classified by an access control list or traffic manager flows.
Abstract:
A network processing system is described that is able to scan the entire contents of data packets passing through it, and to associate related data packets into discrete sessions, or flows. This ability allows the network processing system to learn characteristics of flows and events contained within those flows. Further, the network processing system can remember characteristics and events that have already been learned for use in processing future data packets. And finally, the network processing system can apply treatments to individual data packets and flows based on the characteristics and events learned, as well as previous state that has been maintained for that flow. To accomplish this, the network processing system includes processing engines, referred to as learning state machines, that scan the entire contents of data packets, use specific information in the data packets to associate the data packet with a particular session, or flow, identify characteristics of the data packet and associated flow, as well as events contained within the flow, store a state for the flow that contains the characteristics and events already learned, and determine a treatment for the data packet based on its contents and flow state.
Abstract:
A packet communication system of the present invention has first mode, second mode and third mode to apply to input packets. The first mode is a mode that decides priority of the packet by at least one of the address information and the application information, the second mode is a mode that decides priority of the packet by the DS value, the third mode is a mode that decides rewrite the DS value by at least one of the address information and the application information. A control unit of the packet communication system switches a mode to apply an input packet of the first mode, the second mode and the third mode based on the packet header information of the input packet.
Abstract:
A network transmitter (driver or adaptor/driver combination) reorders packets received for transmission from a higher layer protocol based on packet destinations. The invention reduces bottlenecks at the transmitter and potentially throughout network intermediate system. Destination based reordering may be accomplished through alternative methods, some alternatives taking into account the order at which packets to different destinations are queued by a higher layer protocol.