Variable-length message data processing method and scheduling device

Abstract

The invention provides a variable-length message data processing method and a scheduling device, and belongs to the field of high-speed transmission of data communication. The data processing method provided by the invention comprises the following steps: receiving and storing a variable-length message; querying a route according to the address type information carried by the variable-length message and obtaining output port information; realizing corresponding port scheduling according to the scheduling strategy of the scheduling device; and carrying out data exchange. The switching device executes message caching, sorting, switching and copying operations according to the input port, the type and the destination port of the variable-length message, wherein the variable-length message adopts a shared cache storage and virtual output queue mode in a plane, and the cache space is adjusted according to the number of variable ports. According to the invention, ordered non-blocking exchange of unicast and multicast variable-length messages can be realized according to the port setting of the exchange device.

Description

technical field [0001] The invention belongs to the high-speed transmission field of data communication, and in particular relates to a variable-length message data processing method and a scheduling device. Background technique [0002] PCIE (Peripheral Component Interconnect Express) is a high-speed serial computer expansion bus standard designed to replace the old PCI, PCI-X and AGP bus standards. A typical PCIE topology includes four functional types, root complex (Root Complex), switch (Switch), endpoint (Endpoints) and bridge (bridge). In the field of variable-length message exchange such as PCIeSwitch, variable-length message data exchange mainly involves storage and control information exchange, including data storage exchange, traffic management, and protocol-related sorting. In terms of message data storage, it is mainly stored in the form of cross-bar or Share-memory, and the scheduling method is based on different storage forms. In the cross-bar storage mode, a...

AI Technical Summary

Problems solved by technology

In the cross-bar storage mode, a separate cache node needs to be set up for each input and output combination. The bit width, depth, and number of cache nodes are fixed. The depth of the cache node needs to be calculated according to the line speed requirements and storage delay. For variable-length messages Compared with the exchange of variable ports, the cross-bar mode has a very high rate of waste of storage resources; in the share-memory mode, the cache utilization rate is greatly improved compared with the cross-bar mode, but the shared cache requires multiple For RR (or WRR) or TDM scheduling of port parallel data, it is necessary to increase the system clock frequency or the bit width of the internal data-path to meet the line speed requirements

Images