摘要:
A packet communications system provides for point-to-point packet routing and multicast packet routing to limited subsets of nodes in the network, using a routing field in the packet header which is processed according to two different protocols. A third protocol is provided in which a packet can be multicast to the limited subset even when launched from a node which is not a member of the subset. The routing field includes a first portion which contains the route labels necessary to deliver the packet to the multicast subset. A second portion of the routing field contains the multicast subset identifier which can then be used to deliver the packet to all of the members of the multicast subset. Provision is made to backtrack deliver the packet to the last node identified before the multicast subset if that last node is itself a member of the subset.
摘要:
Method and apparatus for managing internal-node communications in a packet switching network by calculating optimal routes for packets and addressing subnodes within packet nodes using a specific message format. Internal communication facilities called intranode links connect multiple subnodes within nodes. Each subnode contains a switching mechanism and routes packet to other nodes, subnodes, or user applications using a specific message format. The message format allows specific subnodes anywhere in the network to the addressed by any other subnode, making communications more efficient and simplifying the management of internode links.
摘要:
Method and apparatus for making limited internal-node communication facilities externally visible in a packet switching network. Internal-node communication facilities are called intranode links, can include any cable, channel, bus, etc. over which data passes, and are used to connect the multiple subnodes within a given node. Each subnode contains a switching mechanism and routes packets to other nodes, subnodes, or user applications. Each node provides network control functions such as topology, directory, path selection, and bandwidth management which can manage intranode links in the same manner that internode links are currently managed.
摘要:
A method is disclosed for determining whether to use a preempt/resume protocol or an alternate protocol in transmitting data packets from a local system to a remote system. Each system informs the other whether it supports preempt/resume and provides the sizes of the largest low priority data packets it can send and receive. The local system always elects the alternate protocol unless both support preempt/resume. The local system then selects the lesser of (a) the largest low-priority data packet it can send and (b) the largest low-priority data packet the remote system can receive. Use of prompt/resume is initiated by the local system only where a packet of the selected size would be delayed in reaching the remote system by more than a predetermined interval.
摘要:
A device for interconnecting Local Area Networks (LANs) includes ports for attaching LAN segments and port modules for connecting the ports to a switch fabric. Each of the port modules include a mechanism which searches the Routing Information (RI) field of a Received frame to detect at least two Triplets (a minimum configuration for a LAN segment) indicating a Source path from an originator user and a Destination path to a destination user. The Triplet (single or in combination) is used to access a database (tables) which identifies the Port of Exit (POE) through which the frame is to be routed.
摘要:
A LAN switching system includes an Address Match Control line which can be set (activated) and is monitored by each port adapter card. If a port adapter card recognizes an address on the switch fabric, the adapter card copies the frame with the address and activates the Address match Control line. The set Address Match Control line causes the remaining port adapter cards to stop searching for a match. If the Address Match Control line is not set, the frame can be copied by all port adapters which are configured to do so.
摘要:
A dynamic addressing technique mirrors data across multiple banks of a memory resource. Information stored in the memory banks is organized into separately addressable blocks, and memory addresses include a mirror flag. To write information mirrored across two memory banks, a processor issues a single write transaction with the mirror flag asserted. A memory controller detects that the mirror flag is asserted and, in response, waits for both memory banks to become available. At that point, the memory controller causes the write to be performed at both banks. To read data that has been mirrored across two memory banks, the processor issues a read with the mirror flag asserted. The memory controller checks the availability of both banks having the desired information. If either bank is available, the read request is accepted and the desired data is retrieved from the available bank and returned to the processor.
摘要:
A data path protocol eliminates most of the conventional read transactions required to transfer data between devices interconnected by a split transaction bus, such as a HyperTransport (HPT) bus. To that end, each device is configured to manage its own set of buffer descriptors, unlike previous data path protocols in which only one device managed all the buffer descriptors. As such, neither device has to perform a read transaction to retrieve a “free” buffer descriptor from the other device. As a result, only write transactions are performed for transferring descriptors across the HPT bus, thereby decreasing the amount of traffic over the bus and eliminating conventional latencies associated with read transactions. In addition, because descriptors are separately managed in each device, the data path protocol also conserves processing bandwidth that is traditionally consumed by managing ownership of the buffer descriptors within a single device.
摘要:
A buffer-management technique efficiently manages a set of data buffers accessible to first and second devices interconnected by a split transaction bus, such as a Hyper-Transport (HPT) bus. To that end, a buffer manager controls access to a set of “free” buffer descriptors, each free buffer descriptor referencing a corresponding buffer in the set of data buffers. Advantageously, the buffer manager ensures that the first and second devices are allocated a sufficient number of free buffer descriptors for use in a HPT data path protocol in which the first and second devices have access to respective sets of free buffer descriptors. Because buffer management over the HPT bus is optimized by the buffer manager, the amount of processing bandwidth traditionally consumed managing descriptors can be reduced.
摘要:
A backup CRF VLAN arrangement provides an alternate, redundant path for traffic between undistributed Concentrator Relay Functions (CRFs) located on separate switches interconnected by trunk links of a distributed token ring bridge. The backup CRF virtual local area network (VLAN) arrangement defines a backup network path which may be utilized if a primary active path is not a valid path to a backup network. Notably, the backup network comprises a special type of CRF that is distributed among the switches, but that has only one port active at any given time.