Granted Patent B2

US 12,432,145 · App. 17/886,026 · Granted Sep 30, 2025

Inventors: Shrijeet Mukherjee (Mountain View, CA); Shimon Muller (Mountain View, CA); Carlo Contavalli (Mountain View, CA); Gurjeet Singh (Mountain View, CA); Ariel Hendel (Mountain View, CA); Rochan Sankar (Mountain View, CA)

Assignee: Enfabrica Corporation

H04L47/115H04L47/12H04L47/6215

Abstract

Claims (49)

1. A method for congestion control using a flow level transmit mechanism, the method comprising:

obtaining, by a source device, a congestion notification packet (CNP) generated based on congestion in a network, the network connecting the source device and a receiving device;

identifying, by the source device and based on the CNP, a transmit queue in the source device that is causing network congestion;

identifying, by the source device, a transmit block corresponding to the identified transmit queue; and

stopping, by the transmit block, the transmit queue for at least one round trip time (RTT), wherein the RTT is a software programmed value and is updated periodically.

2. The method of claim 1 , wherein the CNP includes information for determining a flow to be stopped, and wherein the information includes a hash computed based on connection information.

3. The method of claim 2 , further comprising:

converting the hash into a receive processing engine index, wherein the hash and the receive processing engine index are used for identifying the transmit queue.

4. The method of claim 1 , further comprising:

determining, by the receiving device, that a receive buffer associated with the receiving device is experiencing underruns; and

in response to determining that the receive buffer is experiencing the underruns, automatically generating the CNP by the receiving device.

5. The method of claim 1 , further comprising:

detecting, by the receiving device, an explicit congestion notification (ECN); and

automatically generating the CNP by the receiving device in response to the ECN.

6. The method of claim 1 , further comprising:

determining, by the source device, that a transmit port in the source device is congested; and

in response to determining that the transmit port is congested, automatically generating the CNP by the source device.

7. The method of claim 1 , wherein the CNP includes an exponential backoff time.

8. The method of claim 1 , wherein the CNP is a user datagram protocol packet sent to a reserved destination port of the source device.

9. The method of claim 1 , further comprising:

forwarding, by the source device, the CNP to a congestion notification queue, wherein the congestion notification queue is a dedicated queue optimized to handle shallow, small packets at a high burst rate; and

processing, by the source device, the congestion notification queue at a highest priority that is assigned using priority-based flow control (PFC).

10. The method of claim 1 , further comprising:

selecting, by the source device and based on the CNP, a receive block from a plurality of receive blocks in the source device; and

forwarding a signal by the receive block to stop a flow to the identified transmit block,

wherein stopping the transmit queue by the transmit block is in response to the signal.

11. A server fabric adapter (SFA) communication system comprising:

a source SFA communicatively couplable to a receive SFA, wherein the source SFA is configured to:

obtain a congestion notification packet (CNP) generated based on congestion in a network, the network connecting the source SFA and the receive SFA;

identify a transmit queue from a plurality of transmit queues in the source SFA based on the CNP, wherein the transmit queue originated a flow causing the congestion; and

stop the transmit queue for at least one round trip time (RTT), wherein the RTT is a software programmed value and is updated periodically.

12. The SFA communication system of claim 11 , wherein the CNP includes information for determining the flow to be stopped, and wherein the information includes a hash computed based on connection information.

13. The SFA communication system of claim 12 , wherein the source SFA is further configured to:

convert the hash into a receive processing engine index, wherein the hash and the receive processing engine index are used for identifying the transmit queue.

14. The SFA communication system of claim 11 , wherein the receive SFA is further configured to:

determine that a receive buffer is experiencing underruns; and

in response to determining that the receive buffer is experiencing the underruns, automatically generate the CNP.

15. The SFA communication system of claim 11 , wherein the receive SFA is further configured to:

detect an explicit congestion notification (ECN); and

automatically generate the CNP in response to the ECN.

16. The SFA communication system of claim 11 , wherein the source SFA is further configured to:

determine that a transmit port is congested; and

in response to determining that the transmit port is congested, automatically generate the CNP.

17. The SFA communication system of claim 11 , wherein the CNP includes an exponential backoff time.

18. The SFA communication system of claim 11 , wherein the CNP is a user datagram protocol packet sent to a reserved destination port of the source SFA.

19. The SFA communication system of claim 11 , wherein the source SFA is further configured to forward the CNP to a congestion notification queue, wherein the congestion notification queue is a dedicated queue optimized to handle shallow, small packets at a high burst rate, and wherein the source SFA is further configured to process the congestion notification queue by processing the dedicated queue at a highest priority that is assigned using priority-based flow control (PFC).

20. The SFA communication system of claim 11 , wherein the source SFA is further configured to:

select a receive block from a plurality of receive blocks in the source SFA based on the CNP; and

forward the CNP to a dedicated congestion notification queue of the receive block.

Assignments (1)

Continuity (2)

References Cited (212)