Emulex Corporation

Exten Technologies, Inc.

Harish Kumar Shakamuri

Ashwin Kamath

Michael Enz

Lawrence Howard Rubin

The present subject disclosure provides a switch architecture with data and control path systolic array that can be used for real time data analysis or Artificial Intelligence (AI) learning. A systolic array is described which analyzes the TLPs received by an uplink port and processes the TLPs according to pre-programmed rules. Then the TLP is forwarded to a destination port. The reverse operation is described as well.

Switch with data and control path systolic array

The present subject disclosure provides a PCIe switch architecture with data and control path systolic array that can be used for real time data analysis or Artificial Intelligence (AI) learning. A systolic array is described which analyzes the TLPs received by an uplink port and processes the TLPs according to pre-programmed rules. Then the TLP is forwarded to a destination port. The reverse operation is described as well.

PCIe switch with data and control path systolic array

Methods and systems are provided for enabling existing or legacy network devices to recognize and parse packets defined in accordance with future-defined standards, without having to be re-configured to be compatible with such standards. A transmitting network device may generate packets such that to enable receiving network devices to bypass, when processing the packets, unknown or newly-inserted fields, such as tag headers, in the packets, and to continue processing the remainder of the packet. This may be achieved by, for example, incorporating in the packets, when such unknown or newly-inserted are included in the packets, corresponding indication fields (e.g., tag header type fields) which may indicated that the unknown or newly-inserted are inserted. Further, indication fields may enable a receiving device to skip over these unknown or newly-inserted. For example, each tag header type filed may be associated with a particular tag header length of the corresponding tag header.

Method for parsing network packets having future defined tags

Disclosed herein is a method allowing an existing or legacy network device to recognize and parse packets defined in accordance with future-defined standards without having to be re-configured to be compatible with such standards. In particular, the device can skip past unknown or newly-inserted fields, such as tag headers in a tagged packet, to parse and process the remainder of the packet. By skipping those added fields such as a tag header, the device can accommodate various future-defined standards without incurring additional engineering or design costs or compromising packet processing efficiency.

Growing community of inventors

Austin, TX, United States of America

Harish Kumar Shakamuri