Abstract: Devices, systems, methods, and processes for sustainably operating a plurality of network planes via de-energization and re-energization is described herein. In many network configurations, a plurality of planes exist that allow for more modular connections in a network fabric. Often, these planes are configured such that each plane is not directly connected to another plane. Because of this, various embodiments described herein can evaluate network conditions and determine if there are conditions suitable to de-energize a plane by either directing the plane to enter a lower-power mode, by shutting off the plane, or disconnecting the available power. This de-energization period can be for a period of time or can occur until a triggering event is detected that indicates that the plane should be re-energized. These determinations can be done based on current traffic trends or historical conditions. They may also be heuristic-based or generated via one or more machine-learning processes.

Abstract: Examples including a system described herein implement a novel liquid time constant neural network (LTC-NN) based architecture to recover an underlying model of physical dynamics from real world data. The automatic differentiation property of LTC-NN nodes overcomes problems associated with low sampling rate, the input dependent time constant in the forward pass of the hidden layer of LTC-NN nodes creates a massive search space of implicit physical dynamics, the physics model solver based data reconstruction loss guides the search for the correct set of implicit dynamics, and drop out in dense layer ensures extraction of the sparsest model. Further, to account for perturbation timing error, the LTC-NN based architecture of the system utilizes dense layer nodes to search through input shifts that results in the lowest reconstruction loss.

Abstract: A system and associated methods extend neural architectures such as liquid time constant neural network (LTC-NN) or continuous time recurrent neural networks (CT-RNN) or neural ordinary differential equations (NODE) to obtain advanced neural structures (LTC-NN-MR, CT-RNN-MR, NODE-MR) that can recover model coefficients of a dynamical system under low sampling rate conditions. The forward pass of these advanced neural structures has the same form as bilinear approximations of nonlinear dynamics. Measurements of real data can be used to convert the set of non-linear dynamics to an over-determined system of equations that are linear in terms of the model coefficients.

Abstract: This disclosure describes techniques and mechanisms to enable proactive convergence of endpoints in a network corresponding to virtual machine mobility scenarios. The techniques may be used individually or together to reduce convergence times and improve network security. The techniques may include proactively exchanging IP-MAC binding information between an old location and a new location of a virtual machine; (2) utilizing a route reflector and a convergence efficient flood order to update the location of the virtual machine at the old location to point to the new location; (3) proactively poisoning the route to the virtual machine at the old location; and (4) utilizing an orchestrator to proactively notify leaf(s) of reachability information of the virtual machine prior to or during migration of the virtual machine.

Abstract: Techniques and architecture are described for a method that includes an in network compute (INC) manager receiving from switches of a fat tree configured network, arithmetic logic unit (ALU) capacity of the switches. Based at least in part on the ALU capacity of the switches and bandwidth, the INC manager determines one or more switches within each tier that are capable of supporting the processing units and based at least in part on the determining, the INC manager selects a first switch as a root, wherein the first switch is included within a tier of switches having intermediate tiers of switches located between the tier and the plurality of processing units within the fat tree configured network. The INC manager creates one or more paths of switches within each of the intermediate tiers from the root to the plurality of processing units to provide a constrained disjoint spanning tree of switches.

Abstract: Devices, systems, methods, and processes for sustainably reallocating resources based on within a plurality of computing nodes of a network, such as a managed network are described herein. Each computing node may be configured to transmit infrastructure data to an infrastructure monitor or ecosystem management tool. Additional sustainability data may also be accessed either internally or externally. The infrastructure data and sustainability data may be utilized to generate one or more scores that can be evaluated against each other. These scores may be configured to reflect various conditions or facts about the computing nodes including the overall sustainability. In order to increase sustainability levels, a variety of different resource configurations can be generated and evaluated against each other and the current configuration.

Abstract: In one example, a method includes receiving, at a network node of a network deploying segment routing, a data packet, wherein an IPv6 header of the data packet includes a source specific function associated with a source address of the data packet and a destination specific function associated with a destination address of the data packet; determining, at the network node, whether a source flag in an End node address of the network node is set; upon determining that the source flag is set, extracting, by the network node, the source specific function; and executing, by the network node, the source specific function for the data packet.

Abstract: The present disclosure discloses a data science driven system and associated methodology framework for mining relevant knowledge from input/output traces of a closed-loop control system for glycemic control to aid in actionable feedback for Type 1 diabetes management. The framework takes Continuous Glucose Monitor (CGM) records and insulin pump records from a closed loop control (CLC) system and interprets the CGM records and insulin pump data to provide medically useful information, wherein the closed-loop control system is operable for operation in an “auto mode” (AP) and a “manual mode” (SAP).

Abstract: This disclosure provides compositions and methods for the editing of a SERPINA1 gene with prime editing.

Abstract: This disclosure describes techniques for performing Internet Protocol (IP) switch-based Storage Area Networks (SANs) analytics management. An example method is performed by an IP switch. The method includes identifying information associated with Non-Volatile Memory express (NVMe)/Transport Control Protocol (TCP) Protocol Data Units (PDUs) examined from an Application-Specific Integrated Circuit (ASIC) data path. The information can be stored based on the NVMe/TCP PDUs being included in separate corresponding TCP data segments, according to a handshake protocol. Alternatively or additionally, the information can be stored based on a metadata header being received as a first PDU in a TCP segment. The NVMe/TCP PDUs can be used to extract the information to populate connection and IO tables inside an ASIC. The storage I/O metrics may then be available from the tables.

Abstract: This disclosure describes techniques for performing Internet Protocol (IP) switch-based Storage Area Networks (SANS) analytics management. An example method is performed by an IP switch. The method includes identifying information associated with Non-Volatile Memory express (NVMe)/Transport Control Protocol (TCP) Protocol Data Units (PDUs) examined from an Application-Specific Integrated Circuit (ASIC) data path. The information can be stored based on the NVMe/TCP PDUs being included in separate corresponding TCP data segments, according to a handshake protocol. Alternatively or additionally, the information can be stored based on a metadata header being received as a first PDU in a TCP segment. The NVMe/TCP PDUs can be used to extract the information to populate connection and IO tables inside an ASIC. The storage I/O metrics may then be available from the tables.

Abstract: This disclosure describes techniques for collecting network parameter data for network switches and/or physical servers and provisioning virtual resources of a service on physical servers based on network resource availability. The network parameter data may include network resource availability data, diagnostic constraint data, traffic flow data, etc. The techniques include determining network switches that have an availability of network resources to support a virtual resource on a connected physical server. A scheduler may deploy virtual machines to particular servers based on the network parameter data in lieu of, or in addition to, the server utilization data of the physical servers (e.g., CPU usage, memory usage, etc.). In this way, a virtual resource may be deployed to a physical server that has an availability of the server resources, but also is connected to a network switch with the availability of network resources to support the virtual resource.

Abstract: This disclosure describes techniques and mechanisms to enable proactive convergence of endpoints in a network corresponding to virtual machine mobility scenarios. The techniques may be used individually or together to reduce convergence times and improve network security. The techniques may include proactively exchanging IP-MAC binding information between an old location and a new location of a virtual machine; (2) utilizing a route reflector and a convergence efficient flood order to update the location of the virtual machine at the old location to point to the new location; (3) proactively poisoning the route to the virtual machine at the old location; and (4) utilizing an orchestrator to proactively notify leaf(s) of reachability information of the virtual machine prior to or during migration of the virtual machine.

Abstract: Examples of a simulation framework are provided to evaluate time varying systems using a piecewise linear time invariant simulation (PLIS) approach. The simulation framework can be configured for an artificial pancreas wireless network system that controls blood glucose in Type 1 Diabetes patients with time varying properties such as physiological changes associated with psychological stress and meal patterns.

Abstract: Systems and methods are provided for determining contextual objects related to web content. Web content may include one or more elements, which may be extracted from the web content and analyzed to identify any contextual objects. The web content elements may be hyperlink elements, and contextual objects may be identified based on the hyperlink elements. A dominant entity may be identified for the web content, and contextual objects relating to the dominant entity may be identified. When a request for web content is received, contextual objects may be provided with the requested content. Personalized contextual objects may be selected from the identified contextual objects based on data associated with the request for web content. Selection of a contextual object may result in navigation to a browsing environment associated with the selected contextual object.

Abstract: Example implementations of a neural network implementing a large language model for generating action plans that align with physical system dynamics of a cyber-physical system (CPS) but are also safe for the human users are disclosed. Examples include a physical dynamics coefficient estimator based on a liquid time constant neural network that can derive coefficients of dynamical models with some unmeasured state variables. Further, the model coefficients are then used to train an LLM with prompts embodied with traces from dynamical system and the corresponding model coefficients. When integrated with a contextualized chatbot, feasible and safe plans can be generated to manage external events such as meals for automated insulin delivery systems used by Type 1 Diabetes subjects.

Abstract: A method of creating a root-of-trust (RoT) within a network fabric may include powering on a network interface card (NIC) baseboard management controller (BMC) (NIC BMC), booting up a NIC via the NIC BMC, obtaining an address for the NIC, verifying an identity of the NIC at a fabric trust identity server using a key obtained from a secure vault communicatively coupled to the NIC BMC, verifying with the fabric trust identity server a number of images of a host device residing in the NIC based at least in part on the identity of the NIC being verified, and instructing a platform BMC to boot up the host device based at least in part on the number of images of the host device being verified.

Abstract: A framework includes a system and associated computer-implemented methods for detecting behavioral changes in a dynamical system that can lead to unsafe conditions before an output of the dynamical system violates a safety threshold, especially for dynamical systems with unmodeled inputs and unmodeled dynamics. In particular, the framework aims to detect “unknown-unknown” errors that may be present in a post-deployment model of the dynamical system that may not be anticipated or modellable by its designers, and are often not directly observable through input-output traces. This is achieved by evaluating conformance of post-deployment model coefficients of the post-deployment model with respect to a set of pre-deployment (ideal) model coefficients. The framework can estimate a future time step where the output of the dynamical system is expected to violate a safety violation based on the post-deployment model.

Abstract: A computer-implemented system (“DeepXSOZ”) exploits synergy between deep-learning based spatial features and shallow-learning based expert knowledge encoding to identify Seizure Onset Zones based on spatiotemporal data captured during brain imaging. DeepXSOZ implements an independent component sorting technique that a) reduces expert sorting workload by 7-fold and b) enables the usage of rs-fMRI as a low-cost outpatient pre-surgical screening tool.

Abstract: Devices, systems, methods, and processes for sustainably operating a plurality of network planes via de-energization and re-energization is described herein. In many network configurations, a plurality of planes exist that allow for more modular connections in a network fabric. Often, these planes are configured such that each plane is not directly connected to another plane. Because of this, various embodiments described herein can evaluate network conditions and determine if there are conditions suitable to de-energize a plane by either directing the plane to enter a lower-power mode, by shutting off the plane, or disconnecting the available power. This de-energization period can be for a period of time or can occur until a triggering event is detected that indicates that the plane should be re-energized. These determinations can be done based on current traffic trends or historical conditions. They may also be heuristic-based or generated via one or more machine-learning processes.