Distributed resilient load-balancing for multipath transport protocols

    公开(公告)号:US11570239B2

    公开(公告)日:2023-01-31

    申请号:US16853048

    申请日:2020-04-20

    Abstract: Techniques are described for providing a distributed application load-balancing architecture that supports multipath transport protocol for client devices connecting to an application service. Rather than having client devices generate new network five-tuples for new subflows to the application servers, the techniques described herein include shifting the burden to the application servers to ensure that the new network five-tuples land in the same bucket in the consistent hashing table. The application servers may receive a hashing function utilized by the load balancers to generate the hash of the network five-tuple. By having the application servers generate the hashes, the load balancers are able to continue stateless, low-level processing of the packets to route them to the correct application servers. In this way, additional subflows can be opened for client devices according to a multipath transport protocol while ensuring that the subflows are routed to the correct application server.

    SMART SERVICE DISCOVERY TO INTERCONNECT CLUSTERS HAVING OVERLAPPING IP ADDRESS SPACE

    公开(公告)号:US20230111266A1

    公开(公告)日:2023-04-13

    申请号:US17498333

    申请日:2021-10-11

    Abstract: An approach to establish connections between clusters having overlapping IP address ranges. A method includes receiving, at a service discovery server, from a first node in a first cluster, a service discovery request including a unique name, determining, at the service discovery server, that the unique name resolves to a destination IP address of a second node in a second cluster, determining that the destination IP address overlaps with an IP address range associated with the first cluster, in response to determining that the destination IP address overlaps with the IP address range belonging to the first cluster, configuring a gateway to expect a network connection request from the first node that includes an IP address of the gateway, and sending a service discovery response to the first node, the service discovery response including the IP address of the gateway, but not the destination IP address.

    TWO-LEVEL CACHE ARCHITECTURE FOR LIVE VIDEO STREAMING THROUGH HYBRID ICN

    公开(公告)号:US20220345507A1

    公开(公告)日:2022-10-27

    申请号:US17862755

    申请日:2022-07-12

    Abstract: A method is provided that is performed by a smart network interface card (SmartNIC) that is in communication with a host device that serves as a forwarder in an Hybrid Information-Centric Network (hICN). The method includes storing in a memory of the SmartNIC, mapping information that maps a subset of names of content contained in a content store that is stored in a memory of the host device to corresponding addresses of the content. The method further includes receiving at the SmartNIC, from a requester, an interest that includes a name, and determining whether the name of the interest is included in the mapping information stored in the SmartNIC. The method includes processing the interest based on whether the name of the interest is contained in the mapping information stored in the memory of the SmartNIC.

    DISTRIBUTED RESILIENT LOAD-BALANCING FOR MULTIPATH TRANSPORT PROTOCOLS

    公开(公告)号:US20210329069A1

    公开(公告)日:2021-10-21

    申请号:US16853048

    申请日:2020-04-20

    Abstract: Techniques are described for providing a distributed application load-balancing architecture that supports multipath transport protocol for client devices connecting to an application service. Rather than having client devices generate new network five-tuples for new subflows to the application servers, the techniques described herein include shifting the burden to the application servers to ensure that the new network five-tuples land in the same bucket in the consistent hashing table. The application servers may receive a hashing function utilized by the load balancers to generate the hash of the network five-tuple. By having the application servers generate the hashes, the load balancers are able to continue stateless, low-level processing of the packets to route them to the correct application servers. In this way, additional subflows can be opened for client devices according to a multipath transport protocol while ensuring that the subflows are routed to the correct application server.

    VIRTUAL ROUTER FUNCTION SHARDING FOR SCALABLE MULTI-TENANT ROUTING

    公开(公告)号:US20240406101A1

    公开(公告)日:2024-12-05

    申请号:US18325431

    申请日:2023-05-30

    Abstract: Proposed herein at techniques for scalable network traffic steering in a multi-tenant network. In one aspect, a method includes receiving, at an ingress of a multi-tenant network, a data packet, the data packet including a Virtual Network Identifier (VNI) identifying a corresponding tenant for the data packet in the multi-tenant network, and determining, at the ingress, a corresponding cluster of routers for processing the data packet based on the VNI, wherein the corresponding cluster of routers is one of a plurality of clusters of routers in the multi-tenant network. The method further includes forwarding the data packet to a router in the corresponding cluster of routers for processing.

    Computer network packet transmission timing

    公开(公告)号:US12052332B2

    公开(公告)日:2024-07-30

    申请号:US17162098

    申请日:2021-01-29

    CPC classification number: H04L69/28 H03M13/09 H04J3/0644 H04J3/065 H04L49/9078

    Abstract: Establishing an expected transmit time at which a network interface controller (NIC) is expected to transmit a next packet. Enqueuing, with the NIC and before the expected transmit time, a packet P1 to be transmitted at the expected transmit time. Upon enqueuing P1, incrementing the expected transmit time by an expected transmit duration of P1. Transmitting at the NIC's line rate and timestamping enqueued P1 with its actual transmit time. Adjusting the expected transmit time by a difference between P1's actual transmit and P1's expected transmit time. Requesting, before completion of transmitting P1, to transmit a P2 at time t(P2). Enqueuing, in sequence, zero or more P0, such that the current expected transmit time plus the duration of the transmission of the P0s at the line rate equals t(P2). Transmitting at the line rate each enqueued P0. Upon enqueuing each P0, incrementing, for each P0, the expected transmit time by the expected transmit duration of the P0. Enqueuing P2 for transmission directly following enqueuing the final P0. Transmitting, by the NIC, enqueued P2 at t(P2).

    REDUCING DISTRIBUTED STORAGE OPERATION LATENCY USING SEGMENT ROUTING TECHNIQUES

    公开(公告)号:US20210185124A1

    公开(公告)日:2021-06-17

    申请号:US17132125

    申请日:2020-12-23

    Abstract: Systems, methods, and computer-readable media for reducing distributed storage operation latency using segment routing. In some examples, a method can involve receiving, from a client, a message identifying an intent to store or retrieve data on a distributed storage environment, and sending to the client a segment routing (SR) list identifying storage node candidates for storing or retrieving the data. The method can involve steering a data request from the client through a path defined by the SR list based on a segment routing header (SRH) associated with the request, the SRH being configured to steer the request through the path until a storage node from the storage node candidates accepts the request. The method can further involve sending, to the client device, a response indicating that the storage node has accepted the request and storing or retrieving the data at the storage node that accepted the request.

Patent Agency Ranking