Out-of-band management techniques for networking fabrics are described. In an example embodiment, an apparatus may comprise a packet-switched network interface to deconstruct a packet received via an out-of-band management network and control circuitry to execute an out-of-band management agent, and the out-of-band management agent may be operative to identify a configuration command comprised in the received packet and control an optical circuit-switched network interface based on the configuration command. Other embodiments are described and claimed.

An example node includes a receiver, a switch circuit, and a transmitter. The receiver is configured to receive a first modulated optical signal including a first plurality of optical subcarriers, and supply a plurality of data streams based on the first plurality of optical subcarriers. Each of the data streams is associated with a corresponding one of the plurality of optical subcarriers. The switch circuit is configured to receive the data streams, and supply the data streams to a plurality of switch outputs. The transmitter is configured to receive the data streams, and supply a second modulated optical signal based on the data streams. The second modulated optical signal carries a second plurality of optical subcarriers. Each of the second plurality of optical subcarriers is associated with a corresponding one of the data streams.

In an example method, network traffic transmitted between a plurality of network nodes via a communications network is monitored. Subsets of the network traffic are ranked according to one or more ranking criteria. A mesh network is deployed between the plurality of network nodes based on the ranking of the subsets of the network traffic. The mesh network includes a plurality of network links, where each network link communicatively couples a respective network node from among the plurality of network nodes to another respective network node from among the plurality of network nodes.

An example system includes a network switch and a plurality of server computers communicatively coupled to the first network switch. The network switch includes a first transceiver configured to transmit data according to a first maximum throughput, and each server computer includes a respective second transceiver configured to transmit data according to a second maximum throughput that is less than the first maximum throughput. The network switch is configured to transmit, using the first transceiver according to the first maximum throughput, first data including a plurality of optical subcarriers to each of the server computers. Each of the server computers is configured to receive, using a respective one of the second transceivers, the first data from the network switch, and extract, from the first data, a respective portion of the first data addressed to the server computer.

An example system includes a plurality of network nodes, each including one or more respective first transceivers configured to transmit data according to a first maximum throughput, and one or more respective second transceivers configured to transmit data according to a second maximum throughput that is less than the first maximum throughput. A first network node is configured to transmit, using a respective one of the first transceivers, first data including a plurality of optical subcarriers to two or more second network nodes according to the first maximum throughput, each optical subcarrier being associated with a different one of the two more other network nodes. The two or more second network nodes are configured to receive, using respective ones of the second transceivers, the first data from the first network node.

An example system includes a first network node, a second network node, and a third network node. The first network node is configured to generate a first optical subcarrier representing first data, and transmit the first optical subcarrier to the second network node. The second network node is configured to receive the first optical subcarrier from the first network node, generate a second optical subcarrier representing the first data, where the second optical subcarrier is different from the first optical subcarrier, and transmit the second optical subcarrier to the third network node.

Technologies for dynamically managing resources in disaggregated accelerators include an accelerator. The accelerator includes acceleration circuitry with multiple logic portions, each capable of executing a different workload. Additionally, the accelerator includes communication circuitry to receive a workload to be executed by a logic portion of the accelerator and a dynamic resource allocation logic unit to identify a resource utilization threshold associated with one or more shared resources of the accelerator to be used by a logic portion in the execution of the workload, limit, as a function of the resource utilization threshold, the utilization of the one or more shared resources by the logic portion as the logic portion executes the workload, and subsequently adjust the resource utilization threshold as the workload is executed. Other embodiments are also described and claimed.

The invention relates to a fibre-optic cross-connection system; in particular having spine-leaf topology, having an input side (S1, S2), in particular a spine side, which has one or a plurality (n) of input switches (S1, S2), Each input switch (S1, S2) comprises a plurality of fibre-optic multi-channel transceivers (QSFP S1.1-S1.4; QSFP S2.1-S2.4), each of which has a number of k fibre-optic channels (Tx0-Tx3). The fibre-optic cross-connection system also has an output side (L1-L4); in particular a leaf side, which has a plurality (m) of output switches (L1, L2, L3, L4) which each have a plurality of fibre-optic multi-channel transceivers (QSFP L1.1-L1.2; QSFP L2.1-L2.2; QSFP L3.1-L3.2; QSFP L4.1-L4.2). The fibre-optic channels (Tx0-Tx3) of at least one, in particular every, input-side multi-channel transceiver (QSFP S1.1-S1.4; QSFP S2.1-S2.4) are divided and connected to output-side multi-channel transceivers (QSFP L1.1-L1.2; QSFP L2.1-L2.2; QSFP L3.1-L3.2; QSFP L4.1-L4.2) which are different from one another, in particular belonging to different output switches (L1, L2, L3, L4).

Today's communications require an effective yet scalable way interconnection of data centers and warehouse scale computers (WSCs) whilst operators must provide a significant portion of data center and WSC applications free of charge to users and consumers. At present, data center operators face the requirement to meet exponentially increasing demand for bandwidth without dramatically increasing the cost and power of the infrastructure employed to satisfy this demand. Simultaneously, consumer expectations of download/upload speeds and latency in accessing content provide additional pressure. Accordingly, the inventors provide a number of optical switching fabrics which reduce the latency and microprocessor loading arising from the prior art Internet Protocol multicasting techniques.

Methods, systems, and apparatus, including an apparatus for generating clusters of building blocks of compute nodes using an optical network. In one aspect, a method includes receiving data specifying requested compute nodes for a computing workload. The data specifies a target arrangement of the nodes. A subset of building blocks of a superpod is selected. A logical arrangement of the subset of compute nodes that matches the target arrangement is determined. A workload cluster of compute nodes that includes the subset of the building blocks is generated. For each dimension of the workload cluster, respective routing data for two or more OCS switches for the dimension is configured. One-to-many switches are configured such that a second compute node of each segment of compute nodes is connected to a same OCS switch as a corresponding first compute node of a corresponding segment to which the second compute node is connected.