Abstract: Telemetry information in the form of platform telemetry, virtualization layer telemetry, and application telemetry can be used to estimate power consumption of a software entity, such as a virtual machine, container, application, or network slice. A controller can take various actions based on software entity power consumption information. If a power limit of an integrated circuit component is exceeded, the controller can reduce the power consumption of a software entity or move the software entity to another integrated circuit component to reduce the power consumption of the integrated circuit component. The controller can determine a total software entity power consumption for software entities associated with a user entity and take actions to keep the total software entity power consumption within a power budget.

Abstract: Various systems and methods for implementing intent-driven power management are described herein. A system includes: a power monitoring unit to collect real-time telemetry of a processor on a compute node; and a power level controller to: receive a power intent for execution of an application on the compute node; configure a power level of the processor of the compute node based on the power intent, the processor to execute the application; set an initial execution priority of the application on the compute node based on the power intent; and modify the initial execution priority based on the power intent and the real-time telemetry of the compute node.

Abstract: Various systems and methods for reactive intent-driven end-to-end (E2E) orchestration are described herein. An orchestrator system, includes a processor; and memory to store instructions, which when executed by the processor, cause the system to: receive, at the orchestrator system, an intent-based service level agreement (SLA) for execution of a series of tasks on a plurality of compute nodes; calculate, based on the intent-based SLA, intermediate latency thresholds corresponding to each task of the series of tasks; calculate slack estimates based on the latency thresholds and real-time telemetry of the plurality of compute nodes or real-time telemetry of connections between the plurality of compute nodes; monitor execution of the series of tasks on the plurality of compute nodes; and perform a corrective action in response to determining that the execution of the series of tasks is predicted to exceed one of the intermediate latency thresholds.

Abstract: Examples described herein relate to an accelerator configured to: perform offloaded decompression of multiple frames of data based on a data compression format, wherein the perform offloaded decompression of the multiple frames of data comprises: based on failure to decompress a frame of the multiple frames of the data: indicate, to a requester, device data identifying at least one of: a successfully decompressed frame of the multiple frames of data or an unsuccessfully decompressed frame.

Abstract: Various systems and methods for implementing intent-based orchestration in heterogenous compute platforms are described herein. An orchestration system is configured to: receive, at the orchestration system, a workload request for a workload, the workload request including an intent-based service level objective (SLO); generate rules for resource allocation based on the workload request; generate a deployment plan using the rules for resource allocation and the intent-based SLO; deploy the workload using the deployment plan; monitor performance of the workload using real-time telemetry; and modify the rules for resource allocation and the deployment plan based on the real-time telemetry.

Abstract: In one embodiment, a system comprises a network interface controller comprising circuitry to determine per-flow analytics information for a plurality of packet flows that are received at the network interface controller and to facilitate differential rate processing of a plurality of packet queues for the plurality of packet flows based on the per-flow analytics information.

Abstract: Disclosed herein are systems, devices, and apparatuses for adaptive data curation. The adaptive data curation system may include a processor configured to receive sensor data about a vehicle, wherein the sensor data is associated with a characteristic label that indicates a collection characteristic about the sensor data. The adaptive data curation system determines based on the characteristic label a processing location of a processing device for processing the sensor data, wherein the processing location is associated with a processing capability of processing the sensor data. The adaptive data curation system routes the sensor data and its characteristic label to the processing device to process the sensor data with the processing capability into an enriched dataset.

Abstract: Various systems and methods for implementing computational storage are described herein. An orchestrator system is configured to: receive, at the orchestrator system, a registration package, the registration package including function code, a logical location of input data for the function code, and an event trigger for the function code, the event trigger set to trigger in response to when the input data is modified; interface with a storage service, the storage service to monitor the logical location of the input data and notify a location service when the input data is modified; interface with the location service to obtain a physical location of the input data, the location service to resolve the physical location from the logical location of the input data; and configure the function code to execute near the input data.

Abstract: A processing device includes a plurality of processing cores, a control register, associated with a first processing core of the plurality of processing cores, to store a first base clock frequency value at which the first processing core is to run, and a power management circuit to receive a base clock frequency request comprising a second base clock frequency value, store the second base clock frequency value in the control register to cause the first processing core to run at the second base clock frequency value, and expose the second base clock frequency value on a hardware interface associated with the power management circuit.

Abstract: Technologies for dynamically managing a batch size of packets include a network device. The network device is to receive, into a queue, packets from a remote node to be processed by the network device, determine a throughput provided by the network device while the packets are processed, determine whether the determined throughput satisfies a predefined condition, and adjust a batch size of packets in response to a determination that the determined throughput satisfies a predefined condition. The batch size is indicative of a threshold number of queued packets required to be present in the queue before the queued packets in the queue can be processed by the network device.

Abstract: Technologies for dynamically managing a batch size of packets include a network device. The network device is to receive, into a queue, packets from a remote node to be processed by the network device, determine a throughput provided by the network device while the packets are processed, determine whether the determined throughput satisfies a predefined condition, and adjust a batch size of packets in response to a determination that the determined throughput satisfies a predefined condition. The batch size is indicative of a threshold number of queued packets required to be present in the queue before the queued packets in the queue can be processed by the network device.

Abstract: In one embodiment, a method comprises determining whether an automation recommendation for a network is to be forwarded to a management system of the network, wherein the automation recommendation specifies an action to take with respect to a plurality of platforms of the network and is received from a recommendation system utilizing artificial intelligence to generate the automation recommendation; responsive to a determination to forward the automation recommendation, assessing an impact of the automation recommendation; and performing an action with respect to the automation recommendation based on the impact.

Abstract: Techniques for reassembling fragmented datagrams are disclosed. Packets may be received and classified. Packets of a fragmented datagram may be stored for later reassembly. In the illustrative embodiment, a datagram is reassembled based on an identified class of service associated with the datagram. Additionally or alternatively, in the illustrative embodiment, a replay window of a replay attack detector may be tuned based on hardware performance of a compute device.

Abstract: Various systems and methods for implementing intent-based cluster administration are described herein. An orchestrator system includes: a processor; and memory to store instructions, which when executed by the processor, cause the orchestrator system to: receive, at the orchestrator system, an administrative intent-based service level objective (SLO) for an infrastructure configuration of an infrastructure; map the administrative intent-based SLO to a set of imperative policies; deploy the set of imperative policies to the infrastructure; monitor performance of the infrastructure; detect non-compliance with the set of imperative policies; and modify the administrative intent-based SLO to generate a revised set of imperative policies that cause the performance of the infrastructure to be compliant with the revised set of imperative policies.

Abstract: A performance monitor provides cache miss stall and memory bandwidth usage metric samples to a resource exhaustion detector. The detector can detect the presence of last-level cache and memory bandwidth exhaustion conditions based on the metric samples. If cache miss stalls and memory bandwidth usage are both trending up, the detector reports a memory bandwidth exhaustion condition to a resource controller. If cache miss stalls are trending up and memory bandwidth usage is trending down, the detector reports a last-level cache exhaustion condition to the resource controller. The resource controller can allocate additional last-level cache or memory bandwidth to the processor unit to remediate the resource exhaustion condition. If bandwidth-related metric samples indicate that a processor unit may be overloaded due to receiving high bandwidth traffic, the resource controller can take a traffic rebalancing remedial action.

Abstract: A computing apparatus, including: a hardware platform; and an interworking broker function (IBF) hosted on the hardware platform, the IBF including a translation driver (TD) associated with a legacy network appliance lacking native interoperability with an orchestrator, the IBF configured to: receive from the orchestrator a network function provisioning or configuration command for the legacy network appliance; operate the TD to translate the command to a format consumable by the legacy network appliance; and forward the command to the legacy network appliance.

Abstract: In one embodiment, a system comprises an interface to receive a plurality of packets; and a plurality of processor units to execute a plurality of transmission sub-interfaces, each transmission sub-interface to perform hierarchical quality of service (HQoS) scheduling on a distinct subset of the plurality of packets, wherein each transmission sub-interface is to schedule its subset of the plurality of packets for transmission by a network interface controller by assigning the packets of the subset to a plurality of transmission queues that each correspond to a distinct traffic class.

Abstract: A system comprising an interface to access a network slice power consumption parameter for a network slice comprising a logical network between two endpoints through a plurality of physical computing platforms; and a controller comprising circuitry, the controller to specify operating parameters for a plurality of hardware resources of a first physical computing platform in accordance with the network slice power consumption parameter.

Abstract: Various systems and methods for implementing intent-based orchestration in heterogenous compute platforms are described herein. An orchestration system is configured to: receive, at the orchestration system, a workload request for a workload, the workload request including an intent-based service level objective (SLO); generate rules for resource allocation based on the workload request; generate a deployment plan using the rules for resource allocation and the intent-based SLO; deploy the workload using the deployment plan; monitor performance of the workload using real-time telemetry; and modify the rules for resource allocation and the deployment plan based on the real-time telemetry.

Abstract: A telemetry manager collects in-band telemetry data and out-of-band telemetry data for an application from a plurality of telemetry sources associated with a plurality of different hardware components in a system. The telemetry manager identifies a subset of hardware components in the plurality of hardware components of the system used to handle workloads of the application and correlates the telemetry data associated with the workloads handled by the subset of hardware components. The telemetry manager triggers a reconfiguration of at least one of the application or the subset of hardware components based on the correlation of the telemetry data.