Mechanism For Achieving Ultra-Low Latency Packet Processing At CU-UP

    公开(公告)号:US20230217341A1

    公开(公告)日:2023-07-06

    申请号:US18068532

    申请日:2022-12-20

    CPC classification number: H04W40/02 H04W80/02

    Abstract: In one embodiment, a method is disclosed, comprising: receiving a packet; processing the packet in a Control Unit-User Plane (CU-UP) as a user-space application with kernel-bypass networking; performing packet Input/Output (I/O) by the CU-UP user-space application, wherein the CU-UP performs Internet Protocol (IP) validation on received user-plane packets before processing Packet Data Convergence Protocol (PDCP), Service Data Adaptation Protocol (SDAP) and General Packet Radio Service Tunnelling Protocol (GTPU) protocol stack and performing direct I/O to a Network Interface Controller (NIC) for sending the packet over a network to a Distributed Unit (DU). The method may further comprise using a containerized CU-UP. The method may further comprise using a plurality of worker threads. The method may further comprise using a polling user space networking accelerator framework.

    Internal Service/Function Discovery
    12.
    发明公开

    公开(公告)号:US20230205752A1

    公开(公告)日:2023-06-29

    申请号:US18146358

    申请日:2022-12-23

    Inventor: Mukesh Singhal

    CPC classification number: G06F16/23 G06F9/4881

    Abstract: A method is disclosed for task and container pod startup and registry in a telecommunications core network, the method comprising: in a telecommunications core network microservices architecture, registering a credential for a container and task in a database as a microservice pod comes up; retrieving container information updated in the database once information for the microservice pod is updated; and sending a message to a controller task to register the microservice pod with the database.

    CU-CP High Availability
    13.
    发明公开

    公开(公告)号:US20230199524A1

    公开(公告)日:2023-06-22

    申请号:US18068520

    申请日:2022-12-19

    Inventor: Mukesh Singhal

    CPC classification number: H04W24/02

    Abstract: A method of providing CU-CP high availability is disclosed, comprising: configuring a single CU-CP instance across a first node and a second node, The single CU-CP instance appears as a single node having multiple IPs to all its peer nodes; keeping session information of the single CU-CP instance at the first node and at the second node; and wherein if the first node fails, the second node may be configured to take up calls from the first node.

    REST Client Scaling over HTTP/1.1
    14.
    发明申请

    公开(公告)号:US20250023938A1

    公开(公告)日:2025-01-16

    申请号:US18781783

    申请日:2024-07-23

    Abstract: A wireless network system is described. In some scenarios MME/MSC/SGSN may act as a client which has huge amount of real-time data to be delivered towards server. For this speedy and reliable delivery requirement, a approach has been proposed which will use multiple parallel HTTP connections for delivery and also make sure that the events related to a particular node is delivered in right order. The multiple HTTP connections are used for parallel delivery so that in case one connections fails then it may not impact other existing connections. The number of connections may grow dynamically up to the configured limit depending on the data delivery requirements and then it may dynamically shrink once the load is reduced. A key based approach is proposed where each connected node will be assigned a unique key which will be used for sequencing the messages related to events of this particular node. In this way we can achieve in-order delivery of data which is dependent on each other along with parallel delivery of data which is independent of each other.

    SCTP micro-service high availability in public/private cloud

    公开(公告)号:US12068914B2

    公开(公告)日:2024-08-20

    申请号:US18183155

    申请日:2023-03-13

    CPC classification number: H04L41/0816

    Abstract: In a first embodiment, a method is disclosed of providing Stream Control Transmission Protocol (SCTP) high availability in a cloud, the method comprising: keeping track, by an internal controller of an internal service discover framework, internal changes in a cluster; determining, by the internal controller, if any SCTP pods crash; informing, by the internal controller, remaining SCTP pods about changes in the environment; determining, by the controller, to distribute a number of connections in each remaining SCTP pod; and initiating, by the remaining pods, SCTP connection reestablishment. In a second embodiment, a Stream Control Transmission Protocol (SCTP) microservice system is disclosed comprising: an internal controller; a plurality of SCTP pods in communication with the internal controller; wherein the internal controller keeps track of internal changes in a cluster, determines if any SCTP pods crash; informs remaining SCTP pods about changes in the environment, and determines distribution of a number of connections in each remaining SCTP pod; and wherein the remaining pods initiate SCTP connection reestablishment.

    Rest client scaling over HTTP/1.1
    16.
    发明授权

    公开(公告)号:US12047431B2

    公开(公告)日:2024-07-23

    申请号:US18053372

    申请日:2022-11-07

    CPC classification number: H04L67/02 G06F8/656

    Abstract: A wireless network system is described. In some scenarios MME/MSC/SGSN may act as a client which has huge amount of real-time data to be delivered towards server. For this speedy and reliable delivery requirement, a approach has been proposed which will use multiple parallel HTTP connections for delivery and also make sure that the events related to a particular node is delivered in right order. The multiple HTTP connections are used for parallel delivery so that in case one connections fails then it may not impact other existing connections. The number of connections may grow dynamically up to the configured limit depending on the data delivery requirements and then it may dynamically shrink once the load is reduced. A key based approach is proposed where each connected node will be assigned a unique key which will be used for sequencing the messages related to events of this particular node. In this way we can achieve in-order delivery of data which is dependent on each other along with parallel delivery of data which is independent of each other.

    SCTP Micro-Service High Availability in Public/Private Cloud

    公开(公告)号:US20230291646A1

    公开(公告)日:2023-09-14

    申请号:US18183155

    申请日:2023-03-13

    CPC classification number: H04L65/65 H04L67/10

    Abstract: In a first embodiment, a method is disclosed of providing Stream Control Transmission Protocol (SCTP) high availability in a cloud, the method comprising: keeping track, by an internal controller of an internal service discover framework, internal changes in a cluster; determining, by the internal controller, if any SCTP pods crash; informing, by the internal controller, remaining SCTP pods about changes in the environment; determining, by the controller, to distribute a number of connections in each remaining SCTP pod; and initiating, by the remaining pods, SCTP connection reestablishment. In a second embodiment, a Stream Control Transmission Protocol (SCTP) microservice system is disclosed comprising: an internal controller; a plurality of SCTP pods in communication with the internal controller; wherein the internal controller keeps track of internal changes in a cluster, determines if any SCTP pods crash; informs remaining SCTP pods about changes in the environment, and determines distribution of a number of connections in each remaining SCTP pod; and wherein the remaining pods initiate SCTP connection reestablishment.

    Singleton Micro-Service High Availability
    18.
    发明公开

    公开(公告)号:US20230208704A1

    公开(公告)日:2023-06-29

    申请号:US18148432

    申请日:2022-12-29

    Inventor: Mukesh Singhal

    CPC classification number: H04L41/0668

    Abstract: A method is disclosed for providing singleton micro-service high availability, comprising: identifying, by an internal controller, that a pod/container has crashed; initiating, by the internal controller, a label change for an existing similar pod labeled as standby to active; matching a singleton micro-service selector label criterion; advertising the newly re-labeled existing similar pod as part of a singleton micro-service; and receiving, by the newly relabeled pod in the system, traffic directed on an IP address of the singleton micro-service.

    REST Client Scaling over HTTP/1.1
    19.
    发明公开

    公开(公告)号:US20230142986A1

    公开(公告)日:2023-05-11

    申请号:US18053372

    申请日:2022-11-07

    CPC classification number: H04L67/02

    Abstract: A wireless network system is described. In some scenarios MME/MSC/SGSN may act as a client which has huge amount of real-time data to be delivered towards server. For this speedy and reliable delivery requirement, a approach has been proposed which will use multiple parallel HTTP connections for delivery and also make sure that the events related to a particular node is delivered in right order. The multiple HTTP connections are used for parallel delivery so that in case one connections fails then it may not impact other existing connections. The number of connections may grow dynamically up to the configured limit depending on the data delivery requirements and then it may dynamically shrink once the load is reduced. A key based approach is proposed where each connected node will be assigned a unique key which will be used for sequencing the messages related to events of this particular node. In this way we can achieve in-order delivery of data which is dependent on each other along with parallel delivery of data which is independent of each other.

Patent Agency Ranking