	US 12,341,703 B2
	Remote direct memory access (RDMA) over converged ethernet with congestion reduction for prevention of loss of data
George Madathilparambil George, Bangalore (IN); Susan George, Bangalore (IN); and Mammen Thomas, Seattle, WA (US)
Filed by George Madathilparambil George, Bangalore (IN); Susan George, Bangalore (IN); and Mammen Thomas, Seattle, WA (US)
Filed on Dec. 23, 2024, as Appl. No. 18/991,816.
Application 15/268,729 is a division of application No. 14/120,845, filed on Jul. 1, 2014, granted, now 9,479,442.
Application 18/991,816 is a continuation of application No. 18/900,714, filed on Sep. 28, 2024, granted, now 12,218,848.
Application 18/900,714 is a continuation of application No. 18/648,425, filed on Apr. 28, 2024, granted, now 12,126,537.
Application 18/648,425 is a continuation of application No. 18/600,441, filed on Mar. 8, 2024, granted, now 12,074,801.
Application 18/600,441 is a continuation of application No. 18/201,779, filed on May 25, 2023, granted, now 11,956,154.
Application 18/201,779 is a continuation of application No. 17/834,097, filed on Jun. 7, 2022, granted, now 11,706,148.
Application 17/834,097 is a continuation of application No. 17/062,594, filed on Oct. 4, 2020, granted, now 11,398,985.
Application 17/062,594 is a continuation of application No. 16/132,427, filed on Sep. 16, 2018, granted, now 10,841,227.
Application 16/132,427 is a continuation of application No. 15/268,729, filed on Sep. 19, 2016, granted, now 10,110,498.
Application 14/120,845 is a continuation of application No. 13/385,155, filed on Feb. 6, 2012, granted, now 8,811,400.
Application 13/385,155 is a continuation of application No. 11/505,788, filed on Aug. 18, 2006, granted, now 8,139,574.
Prior Publication US 2025/0133026 A1, Apr. 24, 2025
Int. Cl. H04L 47/24 (2022.01); G06F 13/40 (2006.01); G06F 13/42 (2006.01); H04L 45/74 (2022.01); H04L 47/125 (2022.01); H04L 49/25 (2022.01); H04L 69/32 (2022.01); H04L 69/324 (2022.01)

CPC H04L 47/24 (2013.01) [G06F 13/4022 (2013.01); G06F 13/4282 (2013.01); H04L 45/74 (2013.01); H04L 47/125 (2013.01); H04L 49/25 (2013.01); H04L 69/32 (2013.01); H04L 69/324 (2013.01)]

15 Claims

1. A network system comprising a plurality of interconnected network nodes, the network nodes comprising:

a) a plurality of network switches, distributed within the network, wherein each of the plurality of network switches are configured for interconnecting network nodes in the network and for forwarding data as datalink frames using a converged ethernet protocol;

b) a plurality of end nodes at network nodes with processing and memory interconnected by the network switches;

c) the end nodes enabled to be source nodes or destination nodes in the network;

d) the plurality of network switches comprising at least a first network switch at a first network node connecting to a second network switch at a second network node and a first end node at a third network node;

e) the second network switch at the second network node in the network connected to the first network switch at the first network node and a second end node at a fourth network node;

f) the first end node configured as a source node for a data and the second end node configured as a destination node for the data;

g) the data being stored at the source node in a source memory and

h) the data configured to be transferred over the network and stored in a destination memory at the second end node;

e) the network switches enabled to use per flow congestion control (PFC) to avoid data loss of priority data due to congestion at a loaded queue indicative of congestion at the second network switch;

g) the network switches configured to implement a method for communication and data transfer between the third network node and the fourth network node; the method comprising:

i) the first network switch receiving a first address of a first data link frame comprising the data from the source node;

ii) receiving the first data link frame by the first switch using Remote Direct Memory Access over converged ethernet (ROCE) technology using the first address;

iii) forwarding the first data link frame to the second switch over the connected network;

iv) writing the first data link frame to the destination memory in the fourth network node using remote direct memory access over converged ethernet (RoCE) technology.