Direct Network Access by a Memory Mapped Peripheral Device for Scheduled Data Transfer on the Network

Abstract

A network interface peripheral device (NIP) may include a network interface for communicating with a network, and an interconnect interface for communicating with a processor subsystem. Peripheral data buffers (PDBs) in the NIP may hold data received from and/or distributed to peer peripherals by the NIP, and network data buffers (NDBs) may hold payload data of scheduled data streams transmitted to and/or received from the network by the NIP. A data handler in the NIP may generate the payload data from the data in the PDBs, and store the payload data in the NDBs according to scheduled data handler transmit events. The data handler may obtain the data from the payload data in the NDBs and store the obtained data in the PDBs according to scheduled data handler receive events. The NIP may include a mirrored finite state machine operating at the device level (of a device that may include the NIP) and controlled by a centralized system configuration entity to manage configuration of the NIP and coordinate the internal configuration of the NIP with a network configuration flow.

Description

FIELD OF THE INVENTION[0001]The present invention relates to the field of distributed systems, and more particularly to direct network access by peripheral devices for scheduled data transfers on the network.DESCRIPTION OF THE RELATED ART[0002]In many industrial applications (and others), instruments collect data or information from an environment or unit under test (UUT), and may also analyze and process acquired data. Some instruments provide test stimuli to a UUT. Examples of instruments include oscilloscopes, digital multimeters, pressure sensors, arbitrary waveform generators, digital waveform generators, etc. The information that may be collected by respective instruments includes information describing voltage, resistance, distance, velocity, pressure, oscillation frequency, humidity, and / or temperature, among others. Computer-based instrumentation systems typically include transducers for capturing a physical phenomenon and generating a representative electrical signal, sign...

AI Technical Summary

Benefits of technology

[0008]A broad range of applications are currently driving the need for distributed data acquisition and control. Examples of such applications include infotainment, industrial control, power transmission and / or generation, transportation control, automotive control, avionics, home networks, media converters, machine / structural health monitoring, and real-time testing, just to name a few. Such applications require the same level of synchronization over a network as achieved inside a chassis. Timely transfer of data from sensors to processing units is becoming increasingly important for “closing the loop” over the network, efficient bandwidth utilization of the network, and reducing cycle time (increasing productivity).

[0009]Converged networking technologies (e.g. IEEE 802.1Q with time-sensitive networking features) enable best-effort and scheduled traffic (latency critical) to coexist on the same network. Various mechanisms in existence today enable implementation of network functions on peripherals to transmit and receive packets of data on the network from memory (CPU access) using a schedule and with minimal jitter. However, there are presently no mechanisms defined for peer peripherals to transmit and receive data on a network via a peripheral implementing the network interface and using a schedule. Various embodiments are presented herein of a system and method for implementing direct network access by a memory mapped peripheral device for scheduled data transfers on the network.

Problems solved by technology

However, there are presently no mechanisms defined for peer peripherals to transmit and receive data on a network via a peripheral implementing the network interface and using a schedule.

Images