DATATYPE ENGINE TO SUPPORT HIGH PERFORMANCE COMPUTING

    公开(公告)号:US20250028450A1

    公开(公告)日:2025-01-23

    申请号:US18905441

    申请日:2024-10-03

    Abstract: A method and apparatus are provided for facilitating a datatype engine (DTE) to support high performance computing. A network interface card (NIC) receives, via a message passing interface, a command to read data from a host memory. The NIC determines that the command indicates a first datatype descriptor stored in the NIC. The NIC forms, based on the command, a packet which indicates a base address and a length associated with the data to be read from the host memory and passes the packet to the DTE. The DTE generates a plurality of read requests comprising offsets from the base address and corresponding lengths based on the first datatype descriptor. The DTE passes the plurality of read requests to a direct memory access module, thereby allowing the NIC to access the host memory while eliminating copies of the data on the host during transfer of the command across a network.

    Facilitating fast re-route in high-availability networks

    公开(公告)号:US12206582B2

    公开(公告)日:2025-01-21

    申请号:US17976732

    申请日:2022-10-28

    Abstract: A system for efficiently determining a set of next-hop switches from a switch is provided. During operation, the system can determine the plurality of next-hop switches for an Internet Protocol (IP) address prefix. The system can then store, in an entry of a forwarding data structure of the switch, a list of identifying information indicating the plurality of next-hop switches corresponding to the IP address prefix. The identifying information for the plurality of next-hop switches can be stored in the list in an order of preference for forwarding traffic matching the IP address prefix. Upon receiving a packet with a destination IP address matching the IP address prefix, the system can select the entry from the forwarding data structure for determining a next-hop switch for forwarding the packet. The system can then determine the next-hop switch for the packet from the entry based on the order of preference.

    Power allotment adjustment in a power sourcing equipment (PSE) of a power over ethernet system based on event detection at another PSE

    公开(公告)号:US12206513B2

    公开(公告)日:2025-01-21

    申请号:US18324551

    申请日:2023-05-26

    Abstract: A Power-over-Ethernet (POE) powered device (PD) may be coupled to two power sourcing equipments (PSEs), a PSE and an additional PSE. The PSE may exchange a transport layer protocol communications with the additional PSE. The communications comprising a first communication from the PSE to the additional PSE indicative of a PoE configuration of the PSE and a second communication from the additional PSE to the PSE indicative of a PoE configuration of the additional PSE. The PSE may create a power availability table based on the communications. The PSE may detect occurrence of an event comprising at least one of a change in the power availability table or a change in ability of the PSE to provide power to the PD. On occurrence of the event, the PSE may send additional communications to the additional PSE, requesting to adjust its power allotment for the PD.

    RESONATORS-BASED PROGRAMMABLE OPTICAL NEURAL NETWORKS

    公开(公告)号:US20250021809A1

    公开(公告)日:2025-01-16

    申请号:US18487452

    申请日:2023-10-16

    Abstract: Systems and methods are provided for devices and methods for implementing an optical neural network (ONN) by leveraging resonator structures, such on micro-ring resonators (MRRs). Examples include unit cells configured to perform a linear transformation on optical signals. Each unit cell comprises a plurality of signal mixing components optically coupled to between adjacent waveguides, where each signal mixing component corresponds to a distinct wavelength and is configured to mix optical signals on the adjacent waveguides at the distinct wavelength. Each unit cell also includes a plurality of phase tuning components each corresponding to a distinct wavelength and configured to adjust a phase of a mixed optical signal at the distinct wavelength.

    FINE-GRAINED ROLE-BASED SEGMENTATION IN OVERLAY NETWORK

    公开(公告)号:US20250016091A1

    公开(公告)日:2025-01-09

    申请号:US18347459

    申请日:2023-07-05

    Abstract: A system for facilitating segmentation by a first switch of an overlay tunnel fabric is provided. During operation, the system can receive a route update packet for the fabric. The packet can be based on a control plane that allows the exchange of route information via the tunnel and can include a first media access control (MAC) address learned at a second switch and a first role identifier of a first role. The first role can indicate a level of access granted to a first device associated with the first MAC address. The system can store the first MAC address and the first role identifier in a local address data structure. Upon receiving a packet from the first device, the system can then determine, based on the first role identifier and a first segmentation policy, whether a local device is allowed to receive the packet from the first device.

    Frame burst overlapping basic service set handling

    公开(公告)号:US12192924B2

    公开(公告)日:2025-01-07

    申请号:US17489426

    申请日:2021-09-29

    Abstract: Systems and methods are provided for frame burst overlapping basic service set (OBSS) handling. In various embodiments, the disclosed methods and systems provide for synchronizing access points (APs) in the same channel. Based on the synchronizing, the APs in the same OBSS channel can determine a distribution policy for using frame bursting. Each AP can use frame bursting in accordance with the distribution policy. In various embodiments, the APs can periodically send frame burst synchronization packets. The frame burst synchronization packets can indicate respective numbers of client devices associated with each AP. Based on the frame burst synchronization packets sent by the APs, a total number of client devices using the same channel can be determined. The distribution policy for using frame bursting can be determined based on the total number of client devices.

    PREVENTING HIGH AVAILABILITY PAIRS OF NETWORK DEVICES FROM SENDING OUT DUPLICATE MULTICAST TRAFFIC FLOWS

    公开(公告)号:US20250007742A1

    公开(公告)日:2025-01-02

    申请号:US18345382

    申请日:2023-06-30

    Abstract: A process includes receiving, at a downstream interface of a high availability pair of network devices, a join request. The availability pair of network devices includes a first network device and a second network device. The join request represents a request by a client device to join a multicast group associated with a source address and a multicast group address. The process includes, responsive to the join request, exporting, by the first network device, first static multicast route information from a second interface of the first network device. The first static multicast route information corresponds to a routing of multicast traffic associated with the multicast group through the first network device. The process further includes, responsive to the join message, controlling, by the second network device, exporting of second static multicast route information from the second network device to prevent duplicate multicast traffic associated with the multicast group from being sent by the high availability pair. The second static multicast route information corresponds to a routing of the multicast traffic associated with the multicast group through the second network device.

Patent Agency Ranking