Systems, methods, and computer-readable media are disclosed for an apparatus coupled to a communication bus, where the apparatus includes a queue and a controller to manage operations of the queue. The queue includes a first space to store a first information for a first traffic type, with a first flow class, and for a first virtual channel of communication between a first communicating entity and a second communicating entity. The queue further includes a second space to store a second information for a second traffic type, with a second flow class, and for a second virtual channel of communication between a third communicating entity and a fourth communicating entity. The first traffic type is different from the second traffic type, the first flow class is different from the second flow class, or the first virtual channel is different from the second virtual channel. Other embodiments may be described and/or claimed.

A relay device, or a switch, in a ring network corrects connection information including (i) identification information of respective switches in the network and (ii) a connection order of other switches seen from a subject switch, by storing in a memory a switch ID table of the subject switch, which includes an ID of each of the other switches and a connection order of the other switches seen from the subject switch, for identifying an abnormal section of the network, for example. Each of the switches has a function that (i) generates anew ID for the restarted switch for generating a new ID table and (ii) causes the other switches to generate a respective new switch ID table, when detecting switch ID duplication among the switches or when restarting from a reset.

A relay device, or a switch, in a ring network corrects connection information including (i) identification information of respective switches in the network and (ii) a connection order of other switches seen from a subject switch, by storing in a memory a switch ID table of the subject switch, which includes an ID of each of the other switches and a connection order of the other switches seen from the subject switch, for identifying an abnormal section of the network, for example. Each of the switches has a function that (i) generates anew ID for the restarted switch for generating a new ID table and (ii) causes the other switches to generate a respective new switch ID table, when detecting switch ID duplication among the switches or when restarting from a reset.

Various embodiments are generally directed to techniques for collective operations among compute nodes in a distributed processing set, such as by utilizing ring sets and local sets of the distributed processing set. In some embodiments, a ring set may include a subset of the distributed processing set in which each compute node is connected to a network with a separate router. In various embodiments, a local set may include a subset of the distributed processing set in which each compute node is connected to a network with a common router. In one or more embodiments, each compute node in a distributed processing set may belong to one ring set and one local set.

Various embodiments are generally directed to techniques for collective operations among compute nodes in a distributed processing set, such as by utilizing ring sets and local sets of the distributed processing set. In some embodiments, a ring set may include a subset of the distributed processing set in which each compute node is connected to a network with a separate router. In various embodiments, a local set may include a subset of the distributed processing set in which each compute node is connected to a network with a common router. In one or more embodiments, each compute node in a distributed processing set may belong to one ring set and one local set.

Aspects include receiving, at an input/output (I/O) processor, a transport control word (TCW) that includes an instruction to perform physical port mirroring. It is identified, by the I/O processor, a first port to be mirrored and a second port to perform the mirroring. The second port is a physical port on a host bus adapter (HBA). In response to outbound data being sent to the first port for transmission to a first target device and to the instruction specifying outbound port mirroring, the I/O processor sends a copy of the outbound data to a second target device via the second port. In response to receiving inbound data at the first port and to the instruction specifying inbound port mirroring, a copy of the inbound data is transmitted to the second target device via the second port.

Aspects include receiving, at an input/output (I/O) processor, a transport control word (TCW) that includes an instruction to perform physical port mirroring. It is identified, by the I/O processor, a first port to be mirrored and a second port to perform the mirroring. The second port is a physical port on a host bus adapter (HBA). In response to outbound data being sent to the first port for transmission to a first target device and to the instruction specifying outbound port mirroring, the I/O processor sends a copy of the outbound data to a second target device via the second port. In response to receiving inbound data at the first port and to the instruction specifying inbound port mirroring, a copy of the inbound data is transmitted to the second target device via the second port.

A method for execution in a central controller comprises obtaining an interconnection topology for a plurality of nodes interconnected via hybrid data/power links; obtaining a power distribution map associated with the topology; and causing DC power to be distributed to the nodes via the links according to the power distribution map. Also, such a node in which there is at least one power-receiving port and at least one power-transmitting port, a controller and power switching circuitry. The controller operates based on power drawn from a portion of the DC power received via the power-receiving port. The controller determines a destination of data packets received via any of the ports and outputting those of the received data packets that are not destined for the network device via another one of the ports. The controller also causes the power switching circuitry to output via plural power-transmitting ports respective portions of the received DC power received at the power-receiving port.

A method for execution in a central controller comprises obtaining an interconnection topology for a plurality of nodes interconnected via hybrid data/power links; obtaining a power distribution map associated with the topology; and causing DC power to be distributed to the nodes via the links according to the power distribution map. Also, such a node in which there is at least one power-receiving port and at least one power-transmitting port, a controller and power switching circuitry. The controller operates based on power drawn from a portion of the DC power received via the power-receiving port. The controller determines a destination of data packets received via any of the ports and outputting those of the received data packets that are not destined for the network device via another one of the ports. The controller also causes the power switching circuitry to output via plural power-transmitting ports respective portions of the received DC power received at the power-receiving port.

The disclosed systems and methods provide hyperscalar packet processing. A method includes receiving a plurality of network packets from a plurality of data paths. The method also includes arbitrating, based at least in part on an arbitration policy, the plurality of network packets to a plurality of packet processing blocks comprising one or more full processing blocks and one or more limited processing blocks. The method also includes processing, in parallel, the plurality of network packets via the plurality of packet processing blocks, wherein each of the one or more full processing blocks processes a first quantity of network packets during a clock cycle, and wherein each of the one or more limited processing blocks processes a second quantity of network packets during the clock cycle that is greater than the first quantity of network packets. The method also includes sending the processed network packets through data buses.