Abstract:
In one embodiment, an apparatus comprises a switch fabric, an adaptation processor to append a length field to a received packet, and a switch fabric to use information in the length field to switch variable-sized packets.
Abstract:
A packet switching system capable of ensuring the sequence and continuity of packets and further compensating for delays in transmission is disclosed. Each of two redundant switch sections has a high-priority queue and a low-priority queue for each of output ports. A high-priority output selector selects one of two high-priority queues corresponding to respective ones of the two switch sections to store an output of the selected one into a high-priority output queue. A low-priority output selector selects one of two low-priority queues corresponding to respective ones of the two switch sections to store an output of the selected one into a low-priority output queue. The high-priority and low-priority output selectors are controlled depending on a system switching signal and a packet storing status of each of the high-priority and low-priority queues.
Abstract:
In a method and multi-queue packet scheduling system for managing network packet traffic, each of a plurality of output buffer queues is assigned with a reserved weight in bandwidth sharing of an output link, is associated with a respective service class, and may be optionally assigned with a maximum rate to constrain packet transmission therethrough. The output buffer queues are adapted for enqueueing incoming packets therein according to associated service classes of the incoming packets. A shared bandwidth rate calculator determines in real-time shared bandwidth rates of maximum-rate constrained backlogged output buffer queues with reference to the reserved weights and such that the shared bandwidth rates do not exceed the assigned maximum rates. A packet scheduler transmits outgoing packets from the backlogged output buffer queues in accordance with a schedule that is based on the shared bandwidth rates determined by the shared bandwidth rate calculator.
Abstract:
A buffered crossbar switch is provided with a buffer to port relationship that supports cells and packets of variable size. A novel scheduler is provided that allows for an efficient crossbar switch, where the relationship between the number of internal buffers is less than the number of ports squared. This allows for a switch that can be implemented that requires less buffer memory.
Abstract:
A packet processing integrated circuit chip includes a plurality of input ports configured to receive packets from respective external sources and a plurality of output ports configured to transmit packets to respective external recipients. The chip further includes a packet processor configurable to extract data from payloads of the received packets, to process the extracted data to produce new packets with payloads having formats compatible with data structures of the external recipients, and to convey the new packets to the output ports. The chip may further include a packet switching fabric configured to route selected packets from the input ports to selected ones of the output ports without payload modification.
Abstract:
The interconnecting network for switching data packets, having data and flow control information, comprises a local packet switch element (S1) with local input buffers (I(1,1) . . . I(1,y)) for buffering the incoming data packets, a remote packet switch element (S2) with remote input buffers (I(2,1) . . . I(2,y)) for buffering the incoming data packets, and data lines (L) for interconnecting the local and the remote packet switch elements (S1, S2). The interconnecting network further comprises a local and a remote arbiter (A1, A2) which are connected via control lines (CL) to the input buffers (I(1,1) . . . I(1,y), I(2,1) . . . I(2,y)), and which are formed such that they can provide that the flow control information is transmitted via the data lines (L) and the control lines (CL).
Abstract:
In a method of dynamic queue management for stable packet forwarding and a network processor element therefor, a network processor of a switch/router can stably assign a packet descriptor for packet forwarding of a local area network/wide are network (LAN/WAN) interface. The method comprises the steps of: determining whether there is a corrupted link for the purpose of processing packets for the forwarding; setting free a packet buffer and a descriptor stored in a queue of a port corresponding to the corrupted link; detecting a normal link to number corresponding output ports; and queuing the packets and descriptors corresponding to the packets to a forwarded one of the calculated ports.
Abstract:
A data switching device has inputs for Granted Throughput (GT) and Best Effort (BE) data, outputs, a data switch interconnecting the inputs and outputs, (GT) control means for controlling the (GT) data scheduling and (BE) control means for controlling the (BE) data scheduling. The (GT) and (BE) control means are arranged for a combined control such that the (BE) data scheduling is based on a contention free (GT) scheduling.
Abstract:
In general, in one aspect, the invention features a method, apparatus, and computer-readable media for sending a frame of data from a first channel to a second channel using at least one of m memory buffers for storing a frame, m being at least 2, in which n of the m buffers have an available status and p of the m buffers have an unavailable status, wherein m=n+p. It comprises reserving q of the n buffers having the available status to the first channel; reserving r of the n buffers having the available status to the second channel, wherein q+r≦n; when a frame is received from the first channel, storing the frame in i of the q buffers, wherein 1≦i≦q, and changing status of the i buffers to unavailable; selectively assigning the frame to the second channel based on a number s of the q buffers, wherein s≦q; and wherein if the frame is assigned to the second channel, the frame is sent to the second channel from the i buffers and the status of the i buffers is changed to available; and if the frame is not assigned to the second channel, the frame is discarded and the status of the i buffers is changed to available.
Abstract translation:通常,一方面,本发明的特征在于一种用于使用m个存储器缓冲器中的至少一个来将数据帧从第一通道发送到第二通道的方法,装置和计算机可读介质,用于存储帧,m处于 至少2,其中m个缓冲器中的n个具有可用状态,并且m个缓冲器的p具有不可用状态,其中m = n + p。 它包括将具有可用状态的n个缓冲器的q保留到第一信道; 将具有可用状态的n个缓冲器的r保留到第二信道,其中q + r <= n; 当从第一信道接收到帧时,将帧存储在q个缓冲器的i中,其中1≤i≤Q,并且将i缓冲器的状态改变为不可用; 基于q个缓冲器的数量s,选择性地将帧分配给第二信道,其中s = q; 并且其中如果所述帧被分配给所述第二信道,则将所述帧从所述i个缓冲器发送到所述第二信道,并且所述i个缓冲器的状态被改变为可用的; 并且如果帧未被分配给第二信道,则丢弃该帧并且将i缓冲器的状态改变为可用的。
Abstract:
An end to end forwarding architecture includes a memory hub having a first ingress interface for receiving packets from a source port. The packets have associated ingress flow identifiers. A second ingress interface outputs the packets to a switch fabric. An ingress controller manages how the packets are queued and output to the switch fabric. The same memory hub can be used for both per flow queuing and per Class of Service (CoS) queuing. A similar structure is used on the egress side of the switch fabric. The end to end forwarding architecture separates per flow traffic scheduling operations performed in a traffic manager from the per flow packet storage operations performed by the memory hub.