Abstract: Systems and methods of routing are provided. In the system, one or more processors determine that a packet is to be transmitted to a destination. In one or more aspects of the system, the one or more processors select a next port to be used for transmitting the packet by selecting a set of ports among a plurality of ports based on a static weight configuration associated with each port. The next port may be selected from the set of ports based on a number of hops required to reach the destination from each port and based on an estimated latency from each port to the destination. The one or more processors may then route the packet through the selected next port.

Abstract: Methods, systems, and apparatus, including instructions encoded on storage media, for performing reduction of gradient vectors and similarly structured data that are generated in parallel, for example, on nodes organized in a mesh or torus topology defined by connections in at least two dimension between the nodes. The methods provide parallel computation and communication between nodes in the topology.

Abstract: Disclosure is made of a shared memory switch and methods and system for controlling such. The shared memory switch may allocate cells in a storage array to respective use cases, the use cases including input buffering, output queuing, free cell allocation, and retry buffering. A set of data packets may be stored in the cells allocated to output queuing, wherein each cell allocated to output queuing stores a respective data packet of the set of data packets. A subset of the set of data packets may be transmitted to a destination external to the shared memory switch. The cells storing the subset of data packets may be reallocated to the retry buffering use case, wherein cells allocated to retry buffering use case are a retry buffer.

Abstract: Disclosure is made of a shared memory switch and methods and system for controlling such. The shared memory switch may allocate cells in a storage array to respective use cases, the use cases including input buffering, output queuing, free cell allocation, and retry buffering. A set of data packets may be stored in the cells allocated to output queuing, wherein each cell allocated to output queuing stores a respective data packet of the set of data packets. A subset of the set of data packets may be transmitted to a destination external to the shared memory switch. The cells storing the subset of data packets may be reallocated to the retry buffering use case, wherein cells allocated to retry buffering use case are a retry buffer.

Abstract: Systems and methods of routing are provided. In the system, one or more processors determine that a packet is to be transmitted to a destination. In one or more aspects of the system, the one or more processors select a next port to be used for transmitting the packet by selecting a set of ports among a plurality of ports based on a static weight configuration associated with each port. The next port may be selected from the set of ports based on a number of hops required to reach the destination from each port and based on an estimated latency from each port to the destination. The one or more processors may then route the packet through the selected next port.

Abstract: Systems and methods of routing are provided. In the system, one or more processors determine that a packet is to be transmitted to a destination. In one or more aspects of the system, the one or more processors select a next port to be used for transmitting the packet by selecting a set of ports among a plurality of ports based on a static weight configuration associated with each port. The next port may be selected from the set of ports based on a number of hops required to reach the destination from each port and based on an estimated latency from each port to the destination. The one or more processors may then route the packet through the selected next port.

Abstract: Systems and methods of routing are provided. In the system, one or more processors determine that a packet is to be transmitted to a destination. In one or more aspects of the system, the one or more processors select a next port to be used for transmitting the packet by selecting a set of ports among a plurality of ports based on a static weight configuration associated with each port. The next port may be selected from the set of ports based on a number of hops required to reach the destination from each port and based on an estimated latency from each port to the destination. The one or more processors may then route the packet through the selected next port.

Abstract: Methods, systems, and apparatus, including instructions encoded on storage media, for performing reduction of gradient vectors and similarly structured data that are generated in parallel, for example, on nodes organized in a mesh or torus topology defined by connections in at least two dimension between the nodes. The methods provide parallel computation and communication between nodes in the topology.

Abstract: Methods, systems, and apparatus, including instructions encoded on storage media, for performing reduction of gradient vectors and similarly structured data that are generated in parallel, for example, on nodes organized in a mesh or torus topology defined by connections in at least two dimension between the nodes. The methods provide parallel computation and communication between nodes in the topology.

Abstract: Methods, systems, and apparatus, including instructions encoded on storage media, for performing reduction of gradient vectors and similarly structured data that are generated in parallel, for example, on nodes organized in a mesh or torus topology defined by connections in at least two dimension between the nodes. The methods provide parallel computation and communication between nodes in the topology.

Abstract: A technique includes using circuit switches to selectively couple packet switches of a switch assembly to the port connectors of the assembly.

Abstract: A system can include an optical multiplexer to combine a plurality of optical input signals having respective wavelengths into a wide-channel optical input signal that is provided to an input channel. The system also includes a photonic packet switch comprising a switch core and a plurality of ports defining a switch radix of the photonic packet switch. The input channel and an output channel can be associated with one of the plurality of ports. The photonic packet switch can process the wide-channel optical input signal and can generate a wide-channel optical output signal that is provided to the output channel. The system further includes an optical demultiplexer to separate the wide-channel optical output signal into a plurality of optical output signals having respective wavelengths. The optical multiplexer and the optical demultiplexer can collectively provide the system with a radix greater than the switch radix.

Abstract: A technique includes generating a checkpoint for an application that is executing on a plurality of nodes of a distributed computing system. Forming the checkpoint includes selectively regulating communication of data from the plurality of nodes to a storage subsystem based at least in part on a replication of the data among the nodes.

Abstract: According to an example, nodes request access to a shared resource in a prioritized optical arbitration system. Each node attempts to extract a wavelength corresponding to a priority level selected by the node from a waveguide. A node is granted access to the resource according to priority level, and is to extract from the waveguide the corresponding wavelength.

Abstract: Examples of the present disclosure include methods, devices, and/or systems. Identifying network communication patterns can include analyzing a distributed computer program of a network, estimating virtual network communication traffic based on the analysis, and mapping the virtual network communication traffic to a physical network link. Identifying network communications patterns can also include identifying the network communication pattern and categorizing the physical communication network link based on an estimated communication intensity of the mapped communication traffic and the network communication pattern. Identifying network communication patterns can further include optimizing an energy used by the network based on the categorization.

Abstract: A power management module can select one of a plurality of different operational modes for a hardware component in a computer system based on application performance and total computer system power consumption determined for each of the operational modes.

Abstract: Continuous recording multichannel disk-based data logging apparatus. The apparatus includes a plurality of disk drives and an interface including a plurality of parallel inputs. The interface is responsive to data at the inputs to write the data on an active plurality of the disk drives, at least one disk drive being idle. The interface is responsive to a user request for data on one of the active disk drives to substitute an idle disk drive into the active plurality in place of that one, to read the requested data, and to cause that one to become idle.

Abstract: A technique includes using circuit switches to selectively couple packet switches of a switch assembly to the port connectors of the assembly.

Abstract: Optical interconnect fabrics and optical switches are disclosed. In one aspect, an optical interconnect fabric comprises one or more bundles of optical broadcast buses. Each optical broadcast bus is optically coupled at one end to a node and configured to transmit optical signals generated by the node. The optical fabric also includes a number of optical tap arrays distributed along each bundle of optical broadcast buses. Each optical tap array is configured to divert a portion of the optical power associated with the optical signals carried by a bundle of optical broadcast buses to one of the nodes.

Abstract: A hybrid electro-optical package for an opto-electronic engine. The package includes a carrier substrate, and a package base. Electrical vias and an optical via of the carrier substrate communicate between a back side and a front side of the carrier substrate. The package base is coupled to the carrier substrate by intra-package electrical interconnects. The carrier substrate is to interconnect electrically with an opto-electronic component mounted on its back side, and includes an optical aperture at its front side for communication of optical signals. Similarly, lands disposed at a front side of the package base provide for communication of electrical signals to an integrated circuit and the opto-electronic component. A system and an opto-electronic engine that include the hybrid electro-optical package are also provided.