Various example embodiments for supporting control over fragmentation of packets in communication networks are described. Various example embodiments for supporting control over fragmentation of packets in communication networks may be configured to support control over fragmentation of Internet Protocol (IP) packets. Various example embodiments for supporting control over fragmentation of IP packets in communication networks may be configured to support control over fragmentation of an IP packet based on inclusion of an IP fragmentability header, including information indicative as to whether the IP packet is permitted to be fragmented, within the IP packet. The IP packet may include a header and a payload, where the header includes an IP packet header and the IP fragmentability header including the information indicative as to whether the IP packet is permitted to be fragmented and, optionally, additional information.

Various aspects of the present disclosure generally relate to wired and/or wireless communication. In some aspects, a device may receive a plurality of data packets at a modem of the device. The device may group, at the modem of the device, payloads of a first subset of the plurality of data packets into a container. The device may transfer, to a processor of the device and using the modem, the container via a first interface channel. The device may transfer, to the processor and using the modem, a second subset of the plurality of data packets via a second interface channel. Numerous other aspects are provided.

Technologies for protocol-agnostic network packet segmentation includes determining whether a size of a payload of a network packet to be transmitted by the compute device exceeds a maximum size threshold and segmenting the payload into a plurality of segmented payloads if the size of the payload exceeds the maximum size of threshold. The payload may be segmented based on segmentation metadata associated with the network packet.

Technologies for protocol-agnostic network packet segmentation includes determining whether a size of a payload of a network packet to be transmitted by the compute device exceeds a maximum size threshold and segmenting the payload into a plurality of segmented payloads if the size of the payload exceeds the maximum size of threshold. The payload may be segmented based on segmentation metadata associated with the network packet.

Techniques are described for generating and using an extended Bi-directional Forwarding Detection (BFD) control packet in a network. The extended BFD control packet includes a control message that includes a BFD session information, an identifier associated with the device sending the BFD control packet, and a payload part. The extended BFD control packet may be used to perform packet loss and/or packet delay related measurements.

The method is for the transfer of data of a message subdivided in fragments from a first intermediary electronic processing unit (43) to a second intermediary electronic processing unit (45); before the transfer, the first unit (43) receives the data encapsulated in the payload of data packets of the TCP type from a sender electronic processing unit (41) and decapsulates them; after the transfer, the second unit (45) encapsulates data in the payload of data packets of the TCP type and transmits them to a recipient electronic processing unit (47); the transfer takes place by means of data packets of the UDP type; the first unit (43) also inserts in the payload (32) of UDP packets; a first data field (C1) containing an identifier of a connection between the sender unit (41) and the recipient unit (47), a second data field (C2) containing an identifier of the message to be transferred, and a third data field (C3) containing a number that identifies the position of a fragment within the message to be transferred.

The method is for the transfer of data of a message subdivided in fragments from a first intermediary electronic processing unit (43) to a second intermediary electronic processing unit (45); before the transfer, the first unit (43) receives the data encapsulated in the payload of data packets of the TCP type from a sender electronic processing unit (41) and decapsulates them; after the transfer, the second unit (45) encapsulates data in the payload of data packets of the TCP type and transmits them to a recipient electronic processing unit (47); the transfer takes place by means of data packets of the UDP type; the first unit (43) also inserts in the payload (32) of UDP packets; a first data field (C1) containing an identifier of a connection between the sender unit (41) and the recipient unit (47), a second data field (C2) containing an identifier of the message to be transferred, and a third data field (C3) containing a number that identifies the position of a fragment within the message to be transferred.

This disclosure provides a method, apparatus, and computer-readable medium for wireless communication at a modem, comprising receiving, via an interface with a host, an internet protocol (IP) packet including a first transport protocol header and a first IP header. The IP packet has a size greater than a maximum transport unit (MTU) size allowable for each packet transmitted over a communication link. The modem segments the IP packet into a plurality of segments based on a segment size indicated by a segmentation policy. Each segment includes a respective derived segment transport protocol header and a respective derived segment IP header derived from the IP packet. Each of these derived headers includes at least one field based on the segmentation policy, and each of the segment transport protocol headers includes a checksum for the respective segment. Additionally, the modem transmits the plurality of segments over the communication link.

In some embodiments, a method receives a packet for a flow from a first application in a first workload to a second application in a second workload. The packet includes an inner header that includes layer 4 information for the first application. The method determines if a setting indicates an outer source port in an outer header should be generated using layer 4 information from the inner header. The setting is based on an analysis of packet types in the flow to determine if fragmented packets are sent. When the setting indicates the outer source port in the outer header should be generated using layer 4 information from the inner header, the method generates the outer source port using the layer 4 information for the first application from the inner header. The packet is encapsulated using the outer header, wherein the outer header includes the outer source port.

In some embodiments, a method receives a packet for a flow from a first application in a first workload to a second application in a second workload. The packet includes an inner header that includes layer 4 information for the first application. The method determines if a setting indicates an outer source port in an outer header should be generated using layer 4 information from the inner header. The setting is based on an analysis of packet types in the flow to determine if fragmented packets are sent. When the setting indicates the outer source port in the outer header should be generated using layer 4 information from the inner header, the method generates the outer source port using the layer 4 information for the first application from the inner header. The packet is encapsulated using the outer header, wherein the outer header includes the outer source port.