Abstract: A network device includes a Network Interface Device (NID) and multiple servers. Each server is coupled to the NID via a corresponding PCIe bus. The NID has a network port through which it receives packets. The packets are destined for one of the servers. The NID detects a PCIe congestion condition regarding the PCIe bus to the server. Rather than transferring the packet across the bus, the NID buffers the packet and places a pointer to the packet in an overflow queue. If the level of bus congestion is high, the NID sets the packet's ECN-CE bit. When PCIe bus congestion subsides, the packet passes to the server. The server responds by returning an ACK whose ECE bit is set. The originating TCP endpoint in turn reduces the rate at which it sends data to the destination server, thereby reducing congestion at the PCIe bus interface within the network device.

Abstract: A network device includes a Network Interface Device (NID) and multiple servers. Each server is coupled to the NID via a corresponding PCIe bus. The NID has a network port through which it receives packets. The packets are destined for one of the servers. The NID detects a PCIe congestion condition regarding the PCIe bus to the server. Rather than transferring the packet across the bus, the NID buffers the packet and places a pointer to the packet in an overflow queue. If the level of bus congestion is high, the NID sets the packet's ECN-CE bit. When PCIe bus congestion subsides, the packet passes to the server. The server responds by returning an ACK whose ECE bit is set. The originating TCP endpoint in turn reduces the rate at which it sends data to the destination server, thereby reducing congestion at the PCIe bus interface within the network device.

Abstract: A first switch in a MPLS network receives a plurality of packets. The plurality of packets are part of a pair of flows. The first switch performs a packet prediction learning algorithm on the first plurality of packets and generates packet prediction information. The first switch communicates the packet prediction information to a Network Operation Center (NOC). In response, the NOC communicates the packet prediction information to a second switch within the MPLS network utilizing OpenFlow messaging. In a first example, the NOC communicates a packet prediction control signal to the second switch. In a second example, a packet prediction control signal is not communicated. In the first example, based on the packet prediction control signal, the second switch determines if it will utilize the packet prediction information. In the second example, the second switch independently determines if the packet prediction information is to be used.

Abstract: A first switch in a MPLS network receives a plurality of packets that are part of a pair of flows. The first switch performs a packet prediction learning algorithm on the first plurality of packets and generates packet prediction information that is communicated to a second switch within the MPLS network utilizing an Operations, Administration, and Maintenance (OAM) packet (message). In a first example, the first switch communicates a packet prediction information notification to a Network Operations Center (NOC) that in response communicates a packet prediction control signal to the second switch. In a second example, the first switch does not communicate a packet prediction information notification. In the first example, the second switch utilizes the packet prediction control signal to determine if the packet prediction information is to be utilized. In the second example, second switch independently determines if the packet prediction information is to be used.

Abstract: A flow of packets is communicated through a data center. The data center includes multiple racks, where each rack includes multiple network devices. A group of packets of the flow is received onto an integrated circuit located in a first network device. The integrated circuit includes a neural network. The neural network analyzes the group of packets and in response outputs a neural network output value. The neural network output value is used to determine how the packets of the flow are to be output from a second network device. In one example, each packet of the flow output by the first network device is output along with a tag. The tag is indicative of the neural network output value. The second device uses the tag to determine which output port located on the second device is to be used to output each of the packets.

Abstract: A flow of packets is communicated through a data center. The data center includes multiple racks, where each rack includes multiple network devices. A group of packets of the flow is received onto an integrated circuit located in one of the network devices. The integrated circuit includes a neural network and a flow table. The neural network analyzes the group of packets and in response determines if it is likely that the flow has a particular characteristic. The neural network outputs a neural network output value that indicates if it is likely that the flow has a particular characteristic. The neural network output value, or a value derived from it, is included in a flow entry in the flow table on the integrated circuit. Packets of the flow subsequently received onto the integrated circuit are routed or otherwise processed according to the flow entry associated with the flow.

Abstract: An appliance receives packets that are part of a flow pair, each packet sharing an application protocol. The appliance determines the application protocol of the packets by performing deep packet inspection (DPI) on the packets. Packet sizes are measured and converted into packet size states. Packet size states, packet sequence numbers, and packet flow directions are used to create an application protocol estimation table (APET). The APET is used during normal operation to estimate the application protocol of a flow pair without performing time consuming DPI. The appliance then determines inter-packet intervals between received packets. The inter-packet intervals are converted into inter-packet interval states. The inter-packet interval states and packet sequence numbers are used to create an inter-packet interval prediction table. The appliance then stores an inter-packet interval prediction table for each application protocol.

Abstract: A flow of packets is communicated through a data center including an electrical switch, an optical switch, and multiple racks each including multiple network devices. The optical switch can be controlled to receive packet traffic from a network device via a first optical link and to output that packet traffic to another network device via a second optical link. One network device includes a neural network that analyzes received packets of the flow. The optical switch is controlled to switch based on a result of the analysis performed. In one instance, the optical switch is controlled such that immediately prior to the switching no packet traffic passes from the first optical link and through the optical switch and to the second optical link but such that after the switching packet traffic does pass from the first optical link and through the optical switch and to the second optical link.

Abstract: An appliance receives packets that are part of a flow pair, each packet sharing an application protocol. The appliance determines an estimated application protocol of the packets without performing deep packet inspection on any packets. The estimated application protocol may be determined by using an application protocol estimation table. The appliance then predicts the inter-packet interval between a packet previously received by the appliance and a next packet not yet received by the appliance. The inter-packet interval may be determined by using an inter-packet interval prediction table. The appliance then preloads packet flow data in a cache before the next packet is predicted to arrive at the appliance. Upon receiving the next packet, the packet flow data is preloaded in the cache. This reduces packet processing time by removing waiting periods previously required to cache packet flow data from an external memory after receiving the next packet.

Abstract: A network appliance includes a network processor and several processing units. Packets a flow pair are received onto the network appliance. Without performing deep packet inspection on any packet of the flow pair, the network processor analyzes the flows, estimates therefrom the application protocol used, and determines a predicted future time when the next packet will likely be received. The network processor determines to send the next packet to a selected one of the processing units based in part on the predicted future time. In some cases, the network processor causes a cache of the selected processing unit to be preloaded shortly before the predicted future time. When the next packet is actually received, the packet is directed to the selected processing unit. In this way, packets are directed to processing units within the network appliance based on predicted future packet arrival times without the use of deep packet inspection.

Abstract: An appliance receives packets that are part of a flow pair, each packet sharing an application protocol. The appliance determines the application protocol of the packets by performing deep packet inspection (DPI) on the packets. Packet sizes are measured and converted into packet size states. Packet size states, packet sequence numbers, and packet flow directions are used to create an application protocol estimation table (APET). The APET is used during normal operation to estimate the application protocol of a flow pair without performing time consuming DPI. The appliance then determines inter-packet intervals between received packets. The inter-packet intervals are converted into inter-packet interval states. The inter-packet interval states and packet sequence numbers are used to create an inter-packet interval prediction table. The appliance then stores an inter-packet interval prediction table for each application protocol.

Abstract: A first switch in a MPLS network receives a plurality of packets that are part of a pair of flows. The first switch performs a packet prediction learning algorithm on the first plurality of packets and generates packet prediction information that is communicated to a second switch within the MPLS network utilizing an Operations, Administration, and Maintenance (OAM) packet (message). In a first example, the first switch communicates a packet prediction information notification to a Network Operations Center (NOC) that in response communicates a packet prediction control signal to the second switch. In a second example, the first switch does not communicate a packet prediction information notification. In the first example, the second switch utilizes the packet prediction control signal to determine if the packet prediction information is to be utilized. In the second example, second switch independently determines if the packet prediction information is to be used.

Abstract: A first switch in a MPLS network receives a plurality of packets. The plurality of packets are part of a pair of flows. The first switch performs a packet prediction learning algorithm on the first plurality of packets and generates packet prediction information. The first switch communicates the packet prediction information to a Network Operation Center (NOC). In response, the NOC communicates the packet prediction information to a second switch within the MPLS network utilizing OpenFlow messaging. In a first example, the NOC communicates a packet prediction control signal to the second switch. In a second example, a packet prediction control signal is not communicated. In the first example, based on the packet prediction control signal, the second switch determines if it will utilize the packet prediction information. In the second example, the second switch independently determines if the packet prediction information is to be used.

Abstract: An appliance receives packets that are part of a flow pair, each packet sharing an application protocol. The appliance determines the application protocol of the packets by performing deep packet inspection (DPI) on the packets. Packet sizes are measured and converted into packet size states. Packet size states, packet sequence numbers, and packet flow directions are used to create an application protocol estimation table (APET). The APET is used during normal operation to estimate the application protocol of a flow pair without performing time consuming DPI. The appliance then determines inter-packet intervals between received packets. The inter-packet intervals are converted into inter-packet interval states. The inter-packet interval states and packet sequence numbers are used to create an inter-packet interval prediction table. The appliance then stores an inter-packet interval prediction table for each application protocol.

Abstract: A network appliance includes a network processor and several processing units. Packets a flow pair are received onto the network appliance. Without performing deep packet inspection on any packet of the flow pair, the network processor analyzes the flows, estimates therefrom the application protocol used, and determines a predicted future time when the next packet will likely be received. The network processor determines to send the next packet to a selected one of the processing units based in part on the predicted future time. In some cases, the network processor causes a cache of the selected processing unit to be preloaded shortly before the predicted future time. When the next packet is actually received, the packet is directed to the selected processing unit. In this way, packets are directed to processing units within the network appliance based on predicted future packet arrival times without the use of deep packet inspection.