Patent classifications
H04L49/9068
MULTI-STREAM SCHEDULING FOR TIME SENSITIVE NETWORKING
A network interface device for implementing multi-stream scheduling for time sensitive networking includes direct memory access (DMA) circuitry, comprising: descriptor parsing circuitry to read a packet descriptor from a descriptor cache, wherein the packet descriptor includes at least one scheduling control parameter including: a launch time offset, a gate cycle offset, or a reduction ratio; wherein the packet descriptor is associated with a packet stream having a traffic class; and scheduling circuitry to schedule packets from the packet stream for transmission using the at least one scheduling control parameter.
Terabit-scale network packet processing via flow-level parallelization
In one example, the present disclosure describes a device, computer-readable medium, and method for organizing terabit-scale packet volumes into flows for downstream processing stages. For instance, in one example, a method includes extracting a first flow key from a first data packet, inputting the first flow key into a hash function to obtain a first output value, selecting a first partition in a memory to which to store the first data packet, wherein the first partition is selected based on the first output value, and storing the first data packet to the first partition.
Data transmission
A method and an apparatus for transmitting data and a network device are provided. In an example of the method, after an interface board and a logical channel both corresponding to first data to be sent are determined, when the logical channel has a capability to send the first data, second data is obtained by adding header information of an interface board identifier and a logical channel identifier to the first data, and stored in a buffer corresponding to the logical channel. Next, when the second data is sent, the second data is read from the buffer, the interface board identifier and the logical channel identifier are obtained from the header information of the read second data, the first data is obtained by removing the header information from the read second data, and the obtained first data is sent to the interface board corresponding to the interface board identifier.
EFFICIENT PACKET RE-TRANSMISSION FOR EXPRESS DATA PATHS
Embodiments of the present disclosure relate to determining modification bounds that identify portions of a packet that are safe to modify so that modified portions of the packet may be flushed from a cache to a memory of a network interface card (NIC) of a host system when the entire packet is synchronized from the NIC. A modification bound of the filter may be determined, and a network packet may be received from the NIC. In response to determining that the network packet is to be modified, a portion of the network packet that is safe to modify may be identified based on the modification bound of the filter and modifications may be made thereto. The modified portion of the network packet may be synchronized to the NIC.
ASYMMETRIC NETWORK INFRASTRUCTURE WITH DISTRIBUTED BROADCAST-SELECT SWITCH AND ASYMMETRIC NETWORK INTERFACE CONTROLLER
Network infrastructure systems including asymmetric Distributed Broadcast Select Switches and Asymmetric Network Interface Controllers for implementation in asymmetric networks and more particularly in cluster networks are provided.
Data Processing Method, Network Interface Card, and Server
A data processing method comprising: After receiving an i.sup.th Peripheral Component Interconnect Express (PCIe) packet, a network interface card stores a j.sup.th instruction segment in a j.sup.th storage unit that is in a first storage area. When all n instruction segments of a first send queue entry (SQE) are stored in the first storage area, the network interface card obtains the first SQE, an identifier of a queue pair (QP) to which the first SQE belongs, and a location identifier of the first SQE in the QP according to the instructions in n storage units in the first storage area; the network interface card performs data processing based on the identifier of the QP to which the first SQE belongs and the location identifier of the first SQE in the QP.
METHODS AND SYSTEMS TO OFFLOAD OVERLAY NETWORK PACKET ENCAPSULATION TO HARDWARE
A method for offloading packet encapsulation for an overlay network is provided. The method, at a virtualization software of a host, sends a mapping table of the overlay network to a physical network interface controller (NIC) associated with the host. The mapping table maps the identification of each of a set of virtual machine (VM) of a tenant on the host to an identification of a tunnel on the overlay network. The method, at the virtualization software, receives a packet from a VM of the tenant. The method sends the packet to the physical NIC. The method, at the physical NIC, encapsulates the packet for transmission over the overlay network by using the mapping table. The method of claim also tags the packet by the virtualization software as a packet that requires encapsulation for transmission in the overlay network prior to sending the packet to the physical NIC.
USER-MODE PROTOCOL STACK-BASED NETWORK ISOLATION METHOD AND DEVICE
A user-mode protocol stack-based network isolation method includes: at a bottom-layer network card interface of a user-mode protocol stack, for each network card, adding an isolation space pointer for binding to a network isolation space; when a service application is initialized, configuring a plurality of user-mode protocol stack network isolation spaces with independent protocol stack private tables based on the user-mode protocol stack; for each network card, designating a corresponding user-mode protocol stack network isolation space through the isolation space pointer of each network card; and for service data received from each network card, performing data processing on the service data through protocol stack private tables in a user-mode protocol stack network isolation space corresponding to the network card.
In-band management interface with user space datapath
A method of utilizing the same hardware network interface card (NIC) in a gateway of a datacenter to communicate datacenter tenant packet traffic and packet traffic for a set of applications that execute in the user space of the gateway and utilize a network stack in the kernel space of the gateway. The method sends and receives packets for the datacenter tenant packet traffic through a packet datapath in the user space. The method sends incoming packets from the NIC to the set of applications through the datapath in the user space, a user-kernel transport driver connecting the kernel network stack to the datapath in the user space, and the kernel network stack. The method receives outgoing packets at the NIC from the set of applications through the kernel network stack, the user-kernel transport driver, and the data path in the user space.
COMMUNICATION APPARATUS GENERATING AND ELIMINATING REDUNDANT DATA PACKETS
Apparatuses, systems, and techniques to eliminate redundant data packets. In at least one embodiment, a communication apparatus generates redundant data packets, and sends them in multiple packet streams. In at least one embodiment, a communication apparatus eliminates redundant data packets from received packet streams.