Technologies for balancing throughput across input ports include a network switch. The network switch is to generate, for an arbiter unit in a first stage of a hierarchy of stages of arbiter units, turn data indicative of a set of turns in which to transfer packet data from devices connected to input ports of the arbiter unit. The network switch is also to transfer, with the arbiter unit, the packet data from the devices in the set of turns. Additionally, the network switch is to determine weight data indicative of the number of turns represented in the set and provide the weight data from the arbiter unit in the first stage to another arbiter unit in a subsequent stage to cause the arbiter unit in the subsequent stage to allocate a number of turns for the transfer of the packet data from the arbiter unit in the first stage.

Communication apparatus includes multiple ports configured to serve as ingress and egress ports, such that the ingress ports receive packets from a packet data network for forwarding to respective egress ports. The ports include an egress port configured for connection to a network interface controller (NIC) serving multiple physical computing units, which have different, respective destination addresses and are connected to the NIC by different, respective communication channels. Control and queuing logic is configured to queue the packets that are received from the packet data network for forwarding to the multiple physical computing units in different, respective queues according to the destination addresses, and to arbitrate among the queues so as to convey the packets from the queues via the same egress port to the NIC, for distribution to the multiple physical computing units over the respective communication channels.

A computer-implemented method, data processing system and computer readable medium manages processing of data entries in a data buffer. Data entries are stored in a data buffer. A representation of a two-directional matrix is annotated to mark the data arrivals as a plurality of entries. In addition, the two-directional matrix is annotated to define a valid processing period for the data entries. The data entries in the data buffer are processed by removing the data entries from the data buffer, marked for processing in the two-dimensional matrix during the valid processing period, by annotating the matrix, labeled as A.sub.CS, to define the number of available entries as a cumulative sum along the columns, A-X, such that A.sub.CS=cumsum(A-X), where cumsum is the cumulative column-sum. Finally, the data entries from the data buffer are processed to determine the entries removed from the data buffer as defined by the row-sum of X. The technique has application for implementing a buffer management system for tax loss carry forward calculations.

Communication apparatus includes multiple ports configured to serve as ingress and egress ports, such that the ingress ports receive packets from a packet data network for forwarding to respective egress ports. The ports include an egress port configured for connection to a network interface controller (NIC) serving multiple physical computing units, which have different, respective destination addresses and are connected to the NIC by different, respective communication channels. Control and queuing logic is configured to queue the packets that are received from the packet data network for forwarding to the multiple physical computing units in different, respective queues according to the destination addresses, and to arbitrate among the queues so as to convey the packets from the queues via the same egress port to the NIC, for distribution to the multiple physical computing units over the respective communication channels.

Technologies for balancing throughput across input ports include a network switch. The network switch is to generate, for an arbiter unit in a first stage of a hierarchy of stages of arbiter units, turn data indicative of a set of turns in which to transfer packet data from devices connected to input ports of the arbiter unit. The network switch is also to transfer, with the arbiter unit, the packet data from the devices in the set of turns. Additionally, the network switch is to determine weight data indicative of the number of turns represented in the set and provide the weight data from the arbiter unit in the first stage to another arbiter unit in a subsequent stage to cause the arbiter unit in the subsequent stage to allocate a number of turns for the transfer of the packet data from the arbiter unit in the first stage.

A system for adaptive audio video (AV) stream processing may include at least one processor and a switch device. The switch device may be configured to route AV traffic to the processor, and to receive AV traffic from the processor and provide the AV traffic to a client device via one or more channels. The processor may monitor a transcoder buffer depth and depths of buffers associated with channels over which the AV traffic is being transmitted. The processor may adaptively modify one or more attributes associated with the AV traffic based at least on the monitored buffer depths. For example, the processor may adaptively adjust a bit rate associated with transcoding the AV traffic based at least on the transcoder buffer depth. The processor may utilize the depths of the buffers associated with the channels to adaptively adjust the amount of AV traffic provided for transmission over the channels.

Examples described herein include a device interface; a first set of one or more processing units; and a second set of one or more processing units. In some examples, the first set of one or more processing units are to perform heavy flow detection for packets of a flow and the second set of one or more processing units are to perform processing of packets of a heavy flow. In some examples, the first set of one or more processing units and second set of one or more processing units are different. In some examples, the first set of one or more processing units is to allocate pointers to packets associated with the heavy flow to a first set of one or more queues of a load balancer and the load balancer is to allocate the packets associated with the heavy flow to one or more processing units of the second set of one or more processing units based, at least in part on a packet receive rate of the packets associated with the heavy flow.

A signal processing aggregation device which switches an offload destination accelerator when offloading specific processing of an application to an accelerator for calculation processing includes: an accelerator offload part which performs offloading on the installed accelerator or the accelerator on a remote server side on the basis of a set offload destination; and an accelerator remote offload part which serializes offload data to the accelerator on the remote server side, packetizes it in a predetermined format, and transmits it as packetized data.