Abstract: A tensor traversal engine in a processor system comprising a source memory component and a destination memory component, the tensor traversal engine comprising: a control signal register storing a control signal for a strided data transfer operation from the source memory component to the destination memory component, the control signal comprising an initial source address, an initial destination address, a first source stride length in a first dimension, and a first source stride count in the first dimension; a source address register communicatively coupled to the control signal register; a destination address register communicatively coupled to the control signal register; a first source stride counter communicatively coupled to the control signal register; and control logic communicatively coupled to the control signal register, the source address register, and the first source stride counter.

Abstract: This document describes techniques and systems for autonomous parking in locations without cellular coverage. A system can include a processor that obtains a request to autonomously park in a parking environment. The system can then determine whether parking space information indicating an available space was obtained. In response to determining that parking space information was obtained, the system can autonomously operate the host vehicle to navigate to and park in the available space. The system determines whether a communication system of the host vehicle has a cellular connection available to aid in the navigation. In response to a determination that the cellular connection is not available, the system communicates the location of the host vehicle to the processor via a proxy vehicle with an operable cellular connection. By utilizing a mesh network among vehicles to share their cellular connections, autonomous parking features may remain available in locations without cellular coverage.

Abstract: Application prototyping systems and methods are disclosed. One aspect is a processing method for multiple computing devices that includes identifying resource constraints for the multiple computing devices. Using identified resource constraints, a presentation model having a plurality of modifiable parameters based at least in part based on the resource constraints is created. At least one inference engine supporting neural network processing is used to execute a particular neural network model based at least in part on the presentation model.

Abstract: Application prototyping systems and methods are disclosed. One aspect is a processing method for multiple computing devices that includes identifying resource constraints for the multiple computing devices. Using identified resource constraints, multiple presentation models at least in part based on identified processing metrics are created. In one aspect, the multiple presentation models include multiple processing pipelines configurable for execution on multiple computing devices. An inference engine can be used to provide an execution model for the multiple processing pipelines based at least in part on the multiple presentation models, with the execution model having improved processing metrics as compared to at least one of the multiple presentation models.

Abstract: The invention relates to a portable device for testing agro-dairy based samples. The said device comprises an optical detector based on micro-spectroscopy. With the aid of an adaptor, the device is amenable to holding a colorimetric test strip, a concave shaped holder for a solid or semi-solid substance and/or a cuvette. The invention also describes a system of testing agro-dairy based samples comprising the said portable device and also a method of testing agro-dairy based samples using the said system.

Abstract: Proxy systems and methods for multiprocessing architectures are described. One method includes receiving an inference request and a statistics request from a client computing system. The method may access a load state of each processing device in a subset of processing devices preloaded with the neural network model, and select a target processing device from the subset based on the load states. One aspect includes transmitting the inference request to the target processing device, and monitoring an execution of the inference request by the target processing device based on the neural network model. The method may receive an inference result generated by the target processing device after executing the inference request, and compute an average inference time for the inference request execution based on the monitoring. The method may transmit the inference result and the average inference time to the client computing system.

Abstract: Proxy systems and methods for multiprocessing architectures are described. One method includes receiving a neural network model from a client computing system. System resource availability on a plurality of processing devices may be assessed, and a subset of available processing devices may be selected based on the system resource availability. In one aspect, the neural network model is loaded into each processing device in the subset. The method may include receiving an inference request from the client computing system. A load state of each processing device in the subset may be accessed, and a target processing device from the subset may be selected based on the load states. The inference request may be transmitted to the target processing device.

Abstract: A tensor traversal engine in a processor system comprising a source memory component and a destination memory component, the tensor traversal engine comprising: a control signal register storing a control signal for a strided data transfer operation from the source memory component to the destination memory component, the control signal comprising an initial source address, an initial destination address, a first source stride length in a first dimension, and a first source stride count in the first dimension; a source address register communicatively coupled to the control signal register; a destination address register communicatively coupled to the control signal register; a first source stride counter communicatively coupled to the control signal register; and control logic communicatively coupled to the control signal register, the source address register, and the first source stride counter.

Abstract: A tensor traversal engine in a processor system comprising a source memory component and a destination memory component, the tensor traversal engine comprising: a control signal register storing a control signal for a strided data transfer operation from the source memory component to the destination memory component, the control signal comprising an initial source address, an initial destination address, a first source stride length in a first dimension, and a first source stride count in the first dimension; a source address register communicatively coupled to the control signal register; a destination address register communicatively coupled to the control signal register; a first source stride counter communicatively coupled to the control signal register; and control logic communicatively coupled to the control signal register, the source address register, and the first source stride counter.

Abstract: An information handling system may include a host system comprising a host system processor, a management controller coupled to the host system processor, and an information handling resource coupled to the host system processor and the management controller, the information handling resource including a firmware. The information handling system may be configured to transfer a firmware update package from the host system to the management controller, wherein the firmware update package includes a cryptographic signature; verify, at the management controller, the cryptographic signature; transfer data indicative of the verification from the management controller to the information handling resource; and in response to receiving the data indicative of the verification from the management controller, install, by the information handling resource, the firmware update package.

Abstract: A tensor traversal engine in a processor system comprising a source memory component and a destination memory component, the tensor traversal engine comprising: a control signal register storing a control signal for a strided data transfer operation from the source memory component to the destination memory component, the control signal comprising an initial source address, an initial destination address, a first source stride length in a first dimension, and a first source stride count in the first dimension; a source address register communicatively coupled to the control signal register; a destination address register communicatively coupled to the control signal register; a first source stride counter communicatively coupled to the control signal register; and control logic communicatively coupled to the control signal register, the source address register, and the first source stride counter.

Abstract: A method for scheduling an artificial neural network includes: accessing a processor representation of a multicore processor comprising processor cores, direct memory access cores, and a cost model; and accessing a network structure defining a set of layers. The method also includes, for each layer in the set of layers: generating a graph based on the processor representation, the graph defining compute nodes, data transfer nodes, and edges representing dependencies between the compute nodes and the data transfer nodes; and generating a schedule for the layer based on the graph, the schedule assigning the compute nodes to the processor cores and assigning the data transfer nodes to the direct memory access cores. The method further includes aggregating the schedule for each layer in the set of layers to generate a complete schedule for the artificial neural network.

Abstract: An information handling system may include a host system comprising a host system processor, a management controller coupled to the host system processor, and an information handling resource coupled to the host system processor and the management controller, the information handling resource including a firmware. The information handling system may be configured to transfer a firmware update package from the host system to the management controller, wherein the firmware update package includes a cryptographic signature; verify, at the management controller, the cryptographic signature; transfer data indicative of the verification from the management controller to the information handling resource; and in response to receiving the data indicative of the verification from the management controller, install, by the information handling resource, the firmware update package.

Abstract: The present invention provides for indirect access to public cloud-based computing environments for the purpose of running ad-hoc commands, on-demand, on any public cloud-based computing resource. Messages including the ad-hoc commands are communicated to a cloud-based command queue, which acts as the intermediary between the user (i.e., command requester) and the public cloud-based computing resources. A command executor agent installed on the cloud-based computing resources, polls the command queue for commands designated for the application(s) currently being executed on the cloud-based computing resources. Results of the command are communicated from the cloud-based computing resource to the command queue, which is periodically polled by a monitoring queue in order for the user/command requester to obtain the results of the execution of the command.

Abstract: The present invention provides for indirect access to public cloud-based computing environments for the purpose of running ad-hoc commands, on-demand, on any public cloud-based computing resource. Messages including the ad-hoc commands are communicated to a cloud-based command queue, which acts as the intermediary between the user (i.e., command requester) and the public cloud-based computing resources. A command executor agent installed on the cloud-based computing resources, polls the command queue for commands designated for the application(s) currently being executed on the cloud-based computing resources. Results of the command are communicated from the cloud-based computing resource to the command queue, which is periodically polled by a monitoring queue in order for the user/command requester to obtain the results of the execution of the command.

Abstract: In one embodiment, a method includes receiving an input signal at a local data lane comprising a dynamic entry shift register, the input signal comprising a marker also received at a remote data lane, identifying receipt of the marker in the local data lane, starting a timer and notifying the remote data lane that the marker was found, receiving a marker found status from the remote data lane and saving a value of the timer, calculating a compensated delay for the remote data lane based on the timer value and a number of pipeline stages for the remote data lane, and setting an entry point to the dynamic entry shift register based on the compensated delay to deskew data between the local data lane and the remote data lane.

Abstract: A controller coordinates execution of a set of related processes executed by respective devices in the virtual network, wherein coordinating comprises causing the respective devices to execute the set of related processes; receiving a data set for the set of related processes from the respective devices, comprising receiving operational states of the related processes from the respective devices; reading a previous data set comprising previous operational states of the related processes from the respective devices; processing an update to the previous operational states from the received operational states of the received data set; and aggregating the received operational states of the data set with the previous operational states of the related processes to form aggregated data of updated operational states.

Abstract: In one embodiment, a method includes receiving an input signal at a local data lane comprising a dynamic entry shift register, the input signal comprising a marker also received at a remote data lane, identifying receipt of the marker in the local data lane, starting a timer and notifying the remote data lane that the marker was found, receiving a marker found status from the remote data lane and saving a value of the timer, calculating a compensated delay for the remote data lane based on the timer value and a number of pipeline stages for the remote data lane, and setting an entry point to the dynamic entry shift register based on the compensated delay to deskew data between the local data lane and the remote data lane.

Abstract: In general, techniques are described in which a plurality of network switches automatically configure themselves to operate as a single virtual network switch. A virtual switch is a collection of individual switch devices that operate like as single network switch. As described herein, network switches in a network that are capable of participating in a virtual switch may automatically discover one another. The participating network switches may then elect one of the participating switches as a master switch. The master switch may generate forwarding information and store the forwarding information in the participating switches, including the master switch. The forwarding information causes the participating switches to act like a single network switch.

Abstract: In general, techniques are described for dynamically modifying the extent of logging performed by logging information generators in response to events detected in logging information received by the collector. In some examples, a network device includes one or more processors and a collector executed by the processors to receive a log message that includes logging information from a generator. The network device also includes a rules engine to apply one or more rules that each specify a condition and a corresponding action to the logging information to identify a matching rule, wherein the rules engine, upon identifying a matching rule, executes the action of the matching rule to generate and send a logging modification message to increase an extent to which the generator generates logging information.