In some examples, a system receives, as part of a process to reset a target device, a one-time value (OTV) generated by the target device. The system sends the OTV from the system to an authentication server. The system receives, from the authentication server, a signature derived from the OTV. The system provides the signature to a controller in the target device that triggers the reset of the target device.
The present disclosure proposes an intelligent scheduling method and system for high-throughput use cases in Wi-Fi 7 Multi-Link Operation (MLO) capable devices. The disclosure schedules and prioritizes Multi-Link Multi-Radio Simultaneous Transmission Reception (MLMR-STR) capable devices and focuses on dynamic channel allocation and load balancing for Access Points (APs). Therefore, through the combination of dynamic channel and bandwidth allocation and load balancing in a network system to address specific high throughput use-cases for Wi-Fi 7 capable MLMR-STR devices, the disclosure takes a holistic approach based on network conditions and client capabilities to optimize throughput of the network system.
Systems, methods, and devices for using distributed write records to maintain concurrency for remote data. A client device prepares to update remote data stored in a memory server that is exposed to the client device for remote memory access. Preparing to update the remote data includes allocating a region of the local memory to update the remote data and, in response to the allocation, storing a write record to the region. The write stores the update data in a data field and a location in one or more location fields. The location denotes where the update data of the write record is to be written. Furthermore, the client device, in response to storing the write record, flags the remote data as having a pending update at the location by updating metadata of the remote data.
Systems and methods are provided for implementing an adaptive chunking process with a similarity metric for improved Retrieval-Augmented Generation (RAG). The process may hierarchically and iteratively determine a set of characters in the document that delineate chunks of data and convert the chunks into a vector embeddings. The hierarchical and iterative determination can be based on a chunking procedure that prioritize paragraphs first, sentences, second, and words third relative to specified or calculated boundary limits.
A processing resource may execute a first confidential virtual machine (CVM) in a trusted execution environment (TEE) at a first privilege level. The first CVM may include a guest operating system (OS) kernel stored in a private memory, which is inaccessible via direct memory access. The processing resource may execute a second CVM in the TEE at a different, second privilege level. Based on the second privilege level, the private memory is accessible to the second CVM. An integrity scanning application of the second CVM may access the private memory to determine a current measurement of the guest OS kernel and may determine whether the guest OS kernel is compromised based on a comparison of the current measurement and an initial measurement of the guest OS kernel.
G06F 21/53 - Contrôle des utilisateurs, des programmes ou des dispositifs de préservation de l’intégrité des plates-formes, p. ex. des processeurs, des micrologiciels ou des systèmes d’exploitation au stade de l’exécution du programme, p. ex. intégrité de la pile, débordement de tampon ou prévention d'effacement involontaire de données par exécution dans un environnement restreint, p. ex. "boîte à sable" ou machine virtuelle sécurisée
G06F 9/455 - ÉmulationInterprétationSimulation de logiciel, p. ex. virtualisation ou émulation des moteurs d’exécution d’applications ou de systèmes d’exploitation
6.
PLUGGABLE STORAGE NORMALIZATION FOR ORCHESTRATION AND MANAGEMENT OF HETEROGENEOUS COMPUTING RESOURCES
A system and method for pluggable normalization of storage orchestration are provided. Data from a storage system and a virtualization system is normalized into a normalized data model. A service request for creating a virtual machine and virtual disk is received. A storage provider plugin for the storage system is loaded. The virtual disk is created in the storage system using the storage provider plugin, which normalizes interactions between the normalized data model and storage system. A hypervisor-specific storage configuration is generated using the storage provider plugin, specifying parameters for attaching the virtual disk to the virtual machine. The virtual machine is created in the virtualization system using the hypervisor-specific storage configuration.
G06F 9/455 - ÉmulationInterprétationSimulation de logiciel, p. ex. virtualisation ou émulation des moteurs d’exécution d’applications ou de systèmes d’exploitation
7.
PLUGGABLE NETWORKING NORMALIZATION FOR ORCHESTRATION AND MANAGEMENT OF HETEROGENEOUS COMPUTING RESOURCES
A system and method for pluggable normalization of networking are provided. Data from a network system is normalized into a normalized data model. A service request calling for configuration of a network device in the network system is received. A network provider plugin for the network system is loaded. A provider-specific network configuration for the network device is generated using the network provider plugin and the normalized data model, wherein the network provider plugin normalizes interactions between the normalized data model and the network system. The provider-specific network configuration is applied to the network device.
According to an implementation, a configurable test circuit enables identical host processor motherboards to support Joint Test Action Group (JTAG) debugging across multiple operating configurations. Each motherboard includes a processor socket, a JTAG header, a transceiver coupled between the JTAG header and processor socket, multiplexers coupled between the processor socket and a connector interface, and a control circuit. In a single processor configuration, JTAG signals route from the header through the processor socket and return to the header. In a dual processor configuration, two identical motherboards couple through their connector interfaces, where one processor functions as a bootstrap processor (BSP) while the other operates as an application processor (AP). The control circuit configures the multiplexers to create a debug chain from the BSP motherboard's JTAG header through both processors.
A device may receive key performance indicators (KPIs) corresponding to a plurality of network devices of a network, and may generate features based on the KPIs. The device may select boundaries for a machine learning model based on the KPIs and the features, and may process the KPIs and the features, with the machine learning model and based on the boundaries, to generate steady state data corresponding to the plurality of network devices. The device may receive new KPIs corresponding to a network device to be added to the network, and may generate new features based on the new KPIs. The device may compare the steady state data with the new KPIs and the new features to identify one or more anomalies corresponding to the network device to be added to the network, and may perform one or more actions based on the one or more anomalies.
H04L 41/0631 - Gestion des fautes, des événements, des alarmes ou des notifications en utilisant l’analyse des causes profondesGestion des fautes, des événements, des alarmes ou des notifications en utilisant l’analyse de la corrélation entre les notifications, les alarmes ou les événements en fonction de critères de décision, p. ex. la hiérarchie ou l’analyse temporelle ou arborescente
G06F 11/07 - Réaction à l'apparition d'un défaut, p. ex. tolérance de certains défauts
H04L 43/0817 - Surveillance ou test en fonction de métriques spécifiques, p. ex. la qualité du service [QoS], la consommation d’énergie ou les paramètres environnementaux en vérifiant la disponibilité en vérifiant le fonctionnement
A network device may receive traffic that includes fragments of a packet, and may store the fragments in a hash table, based on tuple hashes of the fragments, until a first fragment of the packet is received by a network processing unit (NPU) of the network device. The NPU of the network device may determine a service offloaded (SOF) session associated with the packet based on tuples included in the first fragment, and may provide the fragments from the hash table toward a destination based on the SOF session. The network device may store an SOF session identifier in the hash table.
In some implementations, a network device configured as a secure sockets layer forward proxy may receive network traffic associated with an identity. The network device may transmit, using mutual transport layer security, a request for a client certificate associated with the identity. The network device may receive, responsive to the request, the client certificate. The network device may receive, based on the client certificate, identity-specific authentication entry information. The network device may selectively transmit further network traffic based on the identity-specific authentication entry information. The network device may delete, based on the client certificate, a session associated with the identity.
One aspect of the instant disclosure may provide a system and method for processing scatter-gather direct memory access (S-G DMA) instructions. During operation, the system may receive an S-G DMA instruction associated with a message and gather instruction context for the S-G DMA instruction. An S-G DMA processor may process the S-G DMA instruction based on the gathered instruction context and determine whether there exists a pending S-G DMA instruction associated with the message. In response to the presence of the pending S-G DMA instruction, the system stores the instruction context in a hot context cache at an address corresponding to the pending S-G DMA instruction. In response to the absence of the pending S-G DMA instruction, the system stores the instruction context in a cold context cache.
G06F 13/28 - Gestion de demandes d'interconnexion ou de transfert pour l'accès au bus d'entrée/sortie utilisant le transfert par rafale, p. ex. acces direct à la mémoire, vol de cycle
G06F 9/30 - Dispositions pour exécuter des instructions machines, p. ex. décodage d'instructions
G06F 9/38 - Exécution simultanée d'instructions, p. ex. pipeline ou lecture en mémoire
G06F 12/122 - Commande de remplacement utilisant des algorithmes de remplacement du type le moins fréquemment utilisé [LFU], p. ex. avec valeur de comptage individuelle
13.
FIELD PROGRAMMABLE GATE ARRAY WITH INTEGRATED ANALOG IN-MEMORY COMPUTING DOT PRODUCT
According to an implementation, a field programmable gate array includes analog dot product engine (DPE) cores integrated alongside configurable logic circuits. Each DPE core includes an array of programmable resistive memory elements arranged in a crossbar configuration that stores matrix weights and performs matrix-vector multiplication in the analog domain. The DPE cores include digital-to-analog converters, analog-to-digital converters, and shift and add circuitry for processing computation results. A memory array coupled to the DPE cores stores input vectors and partial sums, while an interconnect network couples the configurable logic circuits, DPE cores, and memory array. Control circuitry implements pipeline stages comprising read, compute, sum, activation, and write operations to coordinate data flow between components.
G06F 7/544 - Méthodes ou dispositions pour effectuer des calculs en utilisant exclusivement une représentation numérique codée, p. ex. en utilisant une représentation binaire, ternaire, décimale utilisant des dispositifs n'établissant pas de contact, p. ex. tube, dispositif à l'état solideMéthodes ou dispositions pour effectuer des calculs en utilisant exclusivement une représentation numérique codée, p. ex. en utilisant une représentation binaire, ternaire, décimale utilisant des dispositifs non spécifiés pour l'évaluation de fonctions par calcul
Example implementations relate to deduplication operations in a storage system. An example includes receiving a stream of data units to be stored in a persistent storage of a deduplication storage system, and generating a manifest to record a received order of the data units. The example also includes storing the data units in a first container entity group (CEG) object, and storing a first set of metadata in a first container index, where the first set of metadata includes, for each data unit, a plurality of metadata fields regarding the data unit. The example also includes storing a second set of metadata in a header of the first CEG object, where the second set of metadata includes, for each data unit, a subset of the plurality of metadata fields regarding the data unit.
G06F 16/215 - Amélioration de la qualité des donnéesNettoyage des données, p. ex. déduplication, suppression des entrées non valides ou correction des erreurs typographiques
A piece of networking equipment facilitating efficient congestion management is provided. During operation, the equipment can receive, via a network, a plurality of packets that include portions of a data segment sent from a sender device to a receiver device. The equipment can identify, among the plurality of packets, one or more payload packets comprising payload of the data segment, and at least a header packet comprising header information of the data segment and a header-packet indicator. The equipment can determine whether congestion is detected at the receiver device based on a number of sender devices sending packets to the receiver device via the equipment. Upon determining congestion at the receiver device, the equipment can perform flow trimming by forwarding the header packet to the receiver device and dropping a subset of the one or more payload packets.
The present application relates to an interface integration method of integrating AGV job scheduling system and WMS based on data telegram communication, the method including: performing job assignment, job correction, AGV control, AGV status information acquisition, sequence number synchronization, and communication failure detection based on data telegram; a data structure of the data telegram includes a telegram header, an acknowledgement request flag, a sequence number, a source, a destination, a data area, a check bit, and a telegram trailer. According to the method, by defining the job assignment telegram and the job correction telegram, the job interface integration process of job assignment, job status reporting, job correction and the like is achieved; by defining the control telegram, the WMS can conveniently set the control mode of the AGV; by defining the information acquisition telegram, the WMS can acquire the status information of the AGV at any time.
H04L 47/76 - Contrôle d'admissionAllocation des ressources en utilisant l'allocation dynamique des ressources, p. ex. renégociation en cours d'appel sur requête de l'utilisateur ou sur requête du réseau en réponse à des changements dans les conditions du réseau
H04L 47/70 - Contrôle d'admissionAllocation des ressources
17.
Self-adjusting resource rate limiting based on a dynamic token refresh interval
A network device may determine a token refresh interval, and may generate tokens for a rate limiter based on the token refresh interval. The network device may receive packets or requests to be processed by the rate limiter, and may monitor a consumption of the tokens by the packets or requests during a time interval. The network device may adjust the token refresh interval based on the consumption and to generate a new token refresh interval, and may generate new tokens for the rate limiter based on the new token refresh interval.
A system includes a plurality of servers. Each server includes a baseboard management controller, and each baseboard management controller communicates an associated network traffic flow with a central management server. A virtual gateway appliance of the system includes an aggregator engine. The aggregator engine routes the network traffic flows through respective first persistent network connections associated with respective baseboard management controllers. The aggregator engine provides a second persistent network connection with the central management server, and the aggregator engine multiplexes the network traffic flows to route the network traffic flows through the second persistent network connection.
In some examples, a processing resource converts destination role-focused policies to a plurality of source role-focused policies, and selects a collection of source role-focused policies from among the plurality of source role-focused policies based on a source network address and a destination network address in a first data packet. The processing resource installs the selected collection of source role-focused policies in a policy enforcement hardware controller in an ingress network device for a source compute entity, the policy enforcement hardware controller to enforce a source role-focused policy of the selected collection of source role-focused policies at the ingress network device in response to a second data packet received from the source compute entity.
A development environment for managing application automatic deployment for updated modules for applications that may have a large number of modules asynchronously updated by developers that may have limited insight to other developer activity or to the overall scope of the application being deployed. Continuous Integration/Deployment (CI/CD) tools may be used to identify updates to modules. The CI/CD tools cooperatively operate with graphing tools that track module interdependencies and deployment target relevance, and a module registry to facilitate ranking module deployment order to ensure a deployment order satisfying the module deployment dependencies and deployment target relevance.
One aspect of the instant disclosure provides a system and method for implementing Parallel Redundancy Protocol (PRP) in a network. During operation, a network device operating as a source tunnel endpoint may determine whether a received packet is a Parallel Redundancy Protocol (PRP) packet. In response to determining that the received packet is not a PRP packet, the network device may apply Equal-Cost Multi-Path (ECMP)-based load balancing and forwarding the received packet to an ECMP next-hop network device toward a destination tunnel endpoint. In response to determining that the received packet is a PRP packet, forwarding the received packet to a predetermined next-hop network device toward the destination tunnel endpoint.
In certain examples, a method includes obtaining, by a baseboard management controller (BMC) management platform, static information data sets from static information sources, and a dynamic information data set from a dynamic information source; processing, by a common format document generator of the BMC management platform, the static information data sets and the dynamic information data set to obtain common format documents; processing, by an embedding large language model (LLM), the common format documents to obtain vectors representing data included in the common format documents; receiving, at an inference LLM, a natural language query from a user interface of a BMC of a computing device; and generating, by the inference LLM, a natural language response to the natural language query based at least in part on the natural language query and a portion of the plurality of vectors.
A method for performing recovery operations in response to detecting a boot failure event may include detecting, via a controller, a boot failure event associated with a computing device. The method may also involve querying a database to identify a set of recovery operations that may include one or more scripts associated with recovering from the boot failure event. The database may include a plurality of sets of recovery operations associated with a plurality of boot failure events. The method may then involve retrieving an image based on the set of recovery operations and executing the one or more scripts. The one or more scripts may cause the computing device to return to a previous state.
G06F 11/14 - Détection ou correction d'erreur dans les données par redondance dans les opérations, p. ex. en utilisant différentes séquences d'opérations aboutissant au même résultat
G06F 11/07 - Réaction à l'apparition d'un défaut, p. ex. tolérance de certains défauts
A common method is provided for passing user data in a virtual environment. A system creates, by a tool or program that runs on a hypervisor associated with a plurality of virtual machines (VMs), a data file comprising user data associated with deployment of a VM. The system obtains an output International Standard for Organization (ISO) by creating a copy of a template ISO image. The system updates the ISO image based on the data file, the output ISO image corresponding to a storage medium (e.g., a compact disc/digital versatile disc (CD/DVD)) accessible to a data-reading module of the VM. The system deploys the VM based on the updated output ISO image. The system allows access by the VM to the user data by invoking a call to the data-reading module without requiring hypervisor-specific configuration information.
G06F 9/455 - ÉmulationInterprétationSimulation de logiciel, p. ex. virtualisation ou émulation des moteurs d’exécution d’applications ou de systèmes d’exploitation
25.
GENERIC INTEROPERABILITY AMONG MULTICAST PROTOCOLS
A first network device in a first segment of a network can operate as an RP of a first multicast protocol. A respective multicast route in the first segment is established based on the first multicast protocol. Operating as an RP, the first network device can receive a first join request of the first multicast protocol for a multicast group. The first network device can then generate a second join request for the multicast group based on a second multicast protocol and send the second join request to a second network device in an overlay network coupled to a plurality of segments of the network. Subsequently, the first network device can receive multicast traffic of the multicast group from a second segment of the network via the second network device.
A computer-implemented method may include receiving an onboarding request seeking to onboard a computing device with a management system. The method may include receiving a personal identification number (PIN) from the computing device. The PIN may have a plurality of characters, where specific characters of the PIN encode regional endpoint information and cluster information associated with the management system. The method may include extracting the regional endpoint information and cluster information from the specific characters of the received PIN. The method may include routing, based on the extracted regional endpoint and cluster information, the onboarding request to a cluster of the management system based on the extracted regional endpoint information and cluster information. The method may include validating the received PIN. The method may include transmitting, upon successful validation, onboarding information to the computing device to establish a connection between the computing device and the management system.
In certain embodiments, a method includes stopping platform daemons of a standalone network device; creating an ISSU cache; storing forwarding information in the ISSU cache to freeze the forwarding information; installing an updated state database; storing the forwarding information from the ISSU cache in the updated state database; generating a warm boot file for a line card daemon; installing an updated line card daemon; recovering an internal state of a line card to the updated line card daemon using the warm boot file; connecting the updated line card daemon to the updated state database; syncing the updated state database to a non-updated state database; unfreezing the forwarding information by allowing the updated line card daemon to update the forwarding information of the line card using information from the updated state database; updating control plane daemons of the standalone network device; and updating the platform daemons of the standalone network device.
In some examples, a system computes scores for log templates mapped to log entries containing log data, where a score computed for a respective log template of the log templates is based on characteristics of events relating to log entries represented by the respective log template. The system receives a plurality of log entries for a plurality of components in a computing environment, and filters the plurality of log entries to produce a filtered collection of log entries, where the filtering comprises excluding, from the filtered collection of log entries, a log entry of the plurality of log entries that is represented by a log template assigned a score that fails to satisfy a usefulness criterion. The system triggers a remediation action in the computing environment based on the filtered collection of log entries.
A first network device in a first segment of a network can operate as a rendezvous point (RP) of a multicast protocol. The routing information of the first segment is maintained in a first virtual routing and forwarding (VRF). The first network device can receive source information of a multicast group from a second network device of a second segment of the network. The source information is received based on a second VRF storing routing information of the second segment. The first network device can determine whether to import the source information into the first VRF based on a condition. Upon importing the source information, the first network device can receive a packet of the multicast group from the second network device based on the second VRF. The first network device can forward the packet to a third network device in the first segment based on the first VRF.
Techniques are provided for identifying the location of network communication issues between a client device and an endpoint. Specifically, indicators of performance issues (IPI) data, associated with a network communication between a client device and an endpoint, and queue monitoring data, associated with a plurality of intermediate routing devices on the path between the client device and the endpoint, may be used to determine a particular component of the network communication as a cause of a communication issue between the client device and the endpoint. An alert may be generated indicating the particular component that is determined to be a cause of the communication issue.
H04L 41/0631 - Gestion des fautes, des événements, des alarmes ou des notifications en utilisant l’analyse des causes profondesGestion des fautes, des événements, des alarmes ou des notifications en utilisant l’analyse de la corrélation entre les notifications, les alarmes ou les événements en fonction de critères de décision, p. ex. la hiérarchie ou l’analyse temporelle ou arborescente
In some examples, an intrusion detection module includes a battery and a switch having a first state and a second state. The first state corresponds to a removable cover of an electronic device being closed, and the second state corresponds to the removable cover of the electronic device being open. The intrusion detection module includes an electronic chip having a counter, the counter to advance based on the switch transitioning from the first state to the second state to connect power from the battery to the electronic chip. The intrusion detection module includes a connector, where a count of the counter is accessible through the connector to a controller in the electronic device, the count when different from an expected value indicating an occurrence of a physical intrusion of the electronic device.
Methods and systems relate to receiving, at a network device, first packets from a first source and second packets from a second source. The network device decrements an index by a first value for each flit transmitted and increments the index by a second value for each unused segment transmitted in a flit. Selection circuitry of the network device select one or more packets from the first packets or the second packets to output from the network device. The selection is configured to prioritize fairness between the first source and the second source when the index is less than or equal to a threshold and to prioritize bandwidth usage when the index is greater than the threshold.
In certain examples, a method includes receiving, at a cloud management service, telemetry information and other device information from a network device; determining a device function based on the telemetry information and the other device information; determining a location at which the network device is located; selecting a configuration bundle based on the device function and the location; and configuring the network device based on the configuration bundle.
Systems and methods are provided for determining a carbon footprint for unit processes of the product identifier, including for all components and subcomponents associated with the product identifier. If all unit process calculations for all identifiers executed successfully, the system aggregates the calculated carbon footprints to the desired level of detail and returns the results to an interface.
Hybrid quantum and probabilistic methods for sampling and/or optimization problems may include using a probabilistic processing unit (PPU) to perform classical sampling for part of the problem space and using a quantum processing unit (QPU) to perform quantum sampling for another portion of the problem space. A method may include representing the problem as a graph; partitioning with sparsification the graph into subgraphs; performing classical sampling on a larger subgraph by the PPU; and performing quantum sampling on the smaller subgraph by the QPU. A method may include a non-equilibrium non-local Monte Carlo approach, including obtaining a seed solution; finding backbones or frozen variables in a configuration space; performing classical sampling on the backbones or frozen variables by the PPU, inducing a larger Hamming distance exploration; and providing coordinates from the classical sampling to the QPU and performing quantum sampling based on the coordinates, inducing a smaller Hamming distance exploration.
Examples described herein relate to an Access Point (AP) and a method for managing Multiple Basic Service Set Identifier (MBSSID) groups. The AP may receive a configuration of a target Virtual Access Point (VAP) including information about a first encryption algorithm. In response to determining that the first encryption algorithm does not match with an encryption algorithm corresponding to one of a plurality of transmitting VAPs of a plurality of MBSSID groups configured for a radio of the AP, the AP may create an additional MBSSID group for the target VAP. Further, the AP may configure the target VAP as a transmitting VAP for the additional MBSSID group, and transmit a management frame corresponding to the additional MBSSID group comprising security information corresponding to the first encryption algorithm.
H04L 9/32 - Dispositions pour les communications secrètes ou protégéesProtocoles réseaux de sécurité comprenant des moyens pour vérifier l'identité ou l'autorisation d'un utilisateur du système
H04W 12/03 - Protection de la confidentialité, p. ex. par chiffrement
One aspect of the instant disclosure provides a system and method for energy management in a data center. During operation, the energy-management system may monitor performance associated with a plurality of servers running an application in a data center, identify one or more high-temperature servers among the plurality of servers based on the monitored performance, and identify one or more candidate replacement servers within the data center. In response to determining that a trigger condition is met by a high-temperature server, the energy-management system may replace the high-temperature server with a selected candidate replacement server by migrating, while the application is running, virtualized resources from the high-temperature server to the selected candidate replacement server and may place the high-temperature server in a low-power mode.
In some examples, a security system receives session establishment information from a client device connected to a network, the session establishment information to establish a session between the client device and the security system, wherein the security system is remote from the network. The security system receives, from the client device, a token sent to the client device by a network device of the network responsive to the network device detecting the session establishment information. The security system determines whether the token is authorized by the security system for an entity affiliated with the client device. The security system sends, based on the determining, a response indicating whether the token is authorized, wherein the token if authorized indicates that the network is operated by the entity that the electronic device is affiliated with.
Techniques are provided herein for selective deep packet inspection. Deep packet inspection (DPI) is implemented selectively for particular data packets that are associated with one or more policies that include a DPI component, such as a DPI-identified criteria and/or a DPI-dependent action.
A system receives, by a network interface card (NIC), inputs including an instruction to read or write a payload of a message, a tracker state indicating a round of processing, and a datatype descriptor defining organization of the message payload. The system identifies a current context and a processing state for the instruction. If the datatype descriptor indicates a first type, the system: obtains the current context associated with the first type from a host memory or a cache of the NIC; and creates direct memory access (DMA) instructions corresponding to the received instruction by executing operations in a nested loop. If the datatype descriptor indicates a second type, the system: obtains the current context associated with the second type by fetching vector entries from a buffer of the NIC; and creates the DMA instructions corresponding to the received instruction based on addresses and lengths in the vector entries.
G06F 13/28 - Gestion de demandes d'interconnexion ou de transfert pour l'accès au bus d'entrée/sortie utilisant le transfert par rafale, p. ex. acces direct à la mémoire, vol de cycle
G06F 3/06 - Entrée numérique à partir de, ou sortie numérique vers des supports d'enregistrement
G06F 9/30 - Dispositions pour exécuter des instructions machines, p. ex. décodage d'instructions
An axial pump for delivering liquid coolant to cool an electronic device comprises a conduit defining a liquid flow path and an impeller/rotor disposed in the conduit. A stator is disposed outside the conduit to drive rotation of the impeller/rotor about an axis of rotation extending parallel to the liquid flow path. The stator comprises a first stator subassembly and a second stator subassembly disposed on diametrically opposite lateral sides of the conduit. Each stator subassembly comprises: a frame with a bobbin column, a winding wound around the bobbin column, a top stator core portion inserted into a bore of the bobbin column via a top opening in the bore; and a bottom stator core portion inserted into the bore of the bobbin column via a bottom opening in the bore.
Discs for mounting an information processing device to a rail kit of a two-post rack are disclosed. The information processing includes a chassis, spools, and discs. The chassis includes a first peripheral side and a second peripheral side, each extends parallel to one another along a longitudinal axis of the chassis. The spools extend perpendicularly from each of the first and second peripheral sides in a mutually opposite direction along a lateral axis of the chassis. The discs (track-travelling discs) include at least one disc mounted to one spool on the first peripheral side of the chassis and at least another disc mounted to another spool on the second peripheral side of the chassis. Further, each disc is configured to engage with and travel along a track of a mounting rail installed in the two-post rack such that the information processing device is detachably mounted to the rack.
An example method for managing power-save states of a plurality of wireless links between an access point (AP) and a client device operating in a Multi-Link Operation (MLO) mode is presented. The AP identifies a first set of wireless links over which the client device is unresponsive to a first set of data packets and a second set of wireless links over which the client device is responsive to the first set of data packets. Then, the AP retransmits the first set of data packets over the first set of wireless links, and transmits a second set of data packets over the second set of wireless links. Further, in response to determining that a first count of retransmissions over the first set of wireless links has reached a first threshold value, the AP may configure the client device in a dummy power-save state for the first set of wireless links.
Systems, methods, and devices for rightsizing data center resources to applications and consolidating server nodes including receiving a definition of goals for rightsizing and consolidating one or more server nodes of a server system and receiving a specification of a trigger for optimization. In the optimization, finding one or more sets of application runtime configurations for the server system and performing a trend analysis on a first set of the one or more application runtime configurations. In response to the trend analysis, predicting performance with consolidation of the one or more server nodes based at least in part on the trend analysis and confirming that the predicted performance matches the goals to verify that the first set satisfies the goals. In response to the performance matching the goals, implementing, by one or more processing resources, the first set of the application runtime configurations in the server system.
A system identifies, in a program, a region of code annotated with a power profile. The system translates the program based on a compiler or an interpreter. The system inserts, in the translated program based on the annotated region of code, a call to a power-control function corresponding to the power profile. The power profile comprises a power-saving operation to be performed at or before runtime execution of the translated annotated region of code. The system performs the power-saving operation at or before runtime execution of the translated program based on the inserted call to the power-control function.
Systems and methods are provided for reducing or eliminating a cold start period experienced with a first machine learning model by implementing a second machine learning model to generate forecasts during the first model's cold start period. The second machine learning model may be trained on data that is collected absent the new device or application, and generate forecasts that may be anonymized and generalized for several environments.
A riser assembly includes a riser card removably connectable to an expansion slot of a primary system board of an information processing device. The riser card includes one or more riser-card expansion slots configured to receive an expansion card. A riser cage supports the riser card and defines a space to receive one or more expansion cards connected to the one or more riser-card expansion slots. The riser cage comprises a fixed bracket and an adjustable bracket. The fixed bracket is configured to be attached to a chassis of the information processing device, with the riser card attached to the fixed bracket. The adjustable bracket is coupled to the fixed bracket and comprises a support feature to support an expansion card installed in the riser cage. The adjustable bracket is adjustable between multiple configurations by moving the support feature relative to the fixed bracket along at least two dimensions.
An axial pump for delivering liquid coolant to cool an electronic device comprises a conduit defining a liquid flow path and an impeller/rotor disposed in the conduit. A stator outside the conduit drives rotation of the impeller/rotor about an axis of rotation parallel to the liquid flow path. A printed circuit assembly (PCA) supplies driving signals to the stator. A housing is formed from multiple distinct housing portions having seams therebetween. A thermal encapsulant is injected, in liquid form, into the housing, with the encapsulant later curing to solid form to thermally couple the stator and the PCA to the conduit, thereby enabling cooling of the stator and PCA via the liquid flowing though the conduit. An encapsulant retaining wrap is adhered to the housing prior to the injection of the encapsulant, with the wrap retaining the encapsulant within the housing by covering at least some of the seams.
A system and cooling apparatuses. One cooling apparatus comprises an outer plate comprising a plurality of ramps, a spring plate comprising a plurality of spring fingers, wherein a higher density of spring fingers is located in an area of higher heat flux and a lower density of spring fingers are located in an area of lower heat flux and a cold plate comprising a coolant channel.
One aspect of the instant application provides a system and method for improving data transmission reliability. During operation, a network node may encode a first data block using first and second Forward Error Correction (FEC) codes to generate, respectively, first and second sets of parity symbols and transmit, over a data link, the first data block accompanied by the first set of parity symbols to a link partner. In response to receiving, from the link partner, a request for the second set of parity symbols, the network node may transmit the second set of parity symbols to the link partner. In response to receiving, from the link partner, a retransmission request, the network node may retransmit the first data block accompanied by the first set of parity symbols to the link partner.
Systems and methods are provided for generating synthetic workload data based on real workload data using a first machine learning model. In some examples, the system may classify, using a second machine learning model, the synthetic workload data. The classification may help identify and adjust a system configuration setting of a compute device and, in some examples, retrain the second machine learning model.
G06N 3/0442 - Réseaux récurrents, p. ex. réseaux de Hopfield caractérisés par la présence de mémoire ou de portes, p. ex. mémoire longue à court terme [LSTM] ou unités récurrentes à porte [GRU]
52.
Display screen or portion thereof with a transitional graphical user interface having a transitional multiselect component
The systems and method disclosed herein optimize the deregistration of user equipment (UE) from the 5GC. For example, when the UE operates in single-registration mode, for AMF 3GPP access registration with drFlag attribute set to false (or to be absent) in UDM/UDR, a “new attribute”, singleRegIndication, is defined. The AMF sets the singleRegIndication value to “NO_INDICATION” when there is no N26 interface connection to MME. UDM will not instruct HSS to cancel MME (and SGSN/VLR) if there is an old AMF 3GPP registration for which “NO_INDICATION” or “DEREGISTER_SN” is set as the singleRegIndication value (i.e. the old AMF was already registered with the single registration). Otherwise, the AMF sets the singleRegIndication value to “DEREGISTER_SN” when the AMF has established a N26 interface connection to the old MME. UDM can instruct HSS to cancel the MME (and SGSN/VLR).
H04W 8/04 - Enregistrement dans un registre de localisation nominal ou un serveur d'abonnés locaux [HSS Home Subscriber Server]
H04W 60/04 - Rattachement à un réseau, p. ex. enregistrementSuppression du rattachement à un réseau, p. ex. annulation de l'enregistrement utilisant des événements déclenchés
H04W 60/06 - Annulation de l'enregistrement ou détachement
54.
DYNAMIC TOPIC BASED CATEGORIZATION AND RETRIEVAL AUGMENTED GENERATION
Systems and methods are provided for a retrieval-augmented generation (RAG) system that categorizes resources into dominant topics and groups the resources into subsets corresponding to the dominant topics, which can be used to retrieve those resources that are probabilistically relevant to a topic of a query. Examples generate topics using a topic model that analyzes content of external resources and creates data structures based on the topics that groups the external resources according to the topics. Examples map a query to a data structure of the data structures based on a correspondence between an queried topic of the query and a topic of the data structure, retrieve the external resources corresponding to the mapped data structure using an information retrieval algorithm, and feed the external resources and the query into a generative artificial intelligence model that generates an output responsive to the query based on the external resources.
A boilerplate assembly for two-phase immersion cooling of an information processing system comprises a boilerplate and a manifold attached to the boilerplate. The boilerplate comprises a first face and a second face opposite the first face. The first face is thermally coupled with a processor or other component of an information processing system. The second face is configured to, in an immersed state of the system in an immersion cooling tank, contact and facilitate boiling of liquid coolant. The manifold comprises multiple nozzles and is to receive an input stream of the liquid coolant from the pool and expel multiple output streams of the liquid coolant back into the pool via the nozzles. The nozzles direct the output streams towards the second face of the boilerplate. The streams may disrupt a mixed flow region at the second face and thereby enhance the rate of boiling.
The present disclosure provides techniques for reimaging physical hosts from a first hypervisor to a second hypervisor. A method includes designating one of a plurality of physical hosts as a coordinating host, receiving a service request to reimage target hosts, obtaining installation media for the second hypervisor, and installing the second hypervisor on each target host. The installation is performed by streaming the installation media from the coordinating host to each of the target hosts. The method enables efficient transition between different hypervisors across multiple physical hosts in a coordinated manner.
G06F 9/455 - ÉmulationInterprétationSimulation de logiciel, p. ex. virtualisation ou émulation des moteurs d’exécution d’applications ou de systèmes d’exploitation
57.
HOST REIMAGING FOR ORCHESTRATION AND MANAGEMENT OF HETEROGENOUS COMPUTING RESOURCES
The present disclosure provides techniques for reimaging physical hosts from a first hypervisor to a second hypervisor. A method includes designating one of a plurality of physical hosts as a coordinating host, receiving a service request to reimage target hosts, obtaining installation media for the second hypervisor, and installing the second hypervisor on each target host. The installation is performed by streaming the installation media from the coordinating host to each of the target hosts. The method enables efficient transition between different hypervisors across multiple physical hosts in a coordinated manner.
G06F 9/455 - ÉmulationInterprétationSimulation de logiciel, p. ex. virtualisation ou émulation des moteurs d’exécution d’applications ou de systèmes d’exploitation
In some examples, a system receives connectivity information indicating how a plurality of resources of a computing environment are connected. Based on the connectivity information, the system infers a first collection of tags to associate with the plurality of resources. The system receives one or more further collections of tags provided from one or more sources, and the system generates, based on the first collection of tags and the one or more further collections of tags, an output collection of tags. The system performs a management action in the computing environment using the output collection of tags.
According to an implementation, a method and system provide route stability and network survivability in a software-defined wide area network (SD-WAN) when gateway devices lose connectivity to a central orchestrator. Upon detecting a disconnection, a gateway device initiates a graceful restart timer and preserves routing information while attempting reconnection. If reconnection fails during the graceful restart period, the gateway device enters headless mode operation governed by a headless mode timer. During headless mode, the gateway device maintains routes while increasing internal path costs and modifying routing protocol attributes to indicate reduced path preference. When multiple gateway devices lose orchestrator connectivity, uniform cost increases preserve relative path preferences established before disconnection. Routes remain available with modified preferences until orchestrator connectivity resumes.
In some examples, a fabric-connected device receives, from a requester over a fabric, a request for a data object, the request including an object identifier, the object identifier containing a security token and an identifier of the data object. The fabric-connected device validates the security token included in the object identifier, and based on validating the security token, initiates an access operation of the data object.
H04L 9/32 - Dispositions pour les communications secrètes ou protégéesProtocoles réseaux de sécurité comprenant des moyens pour vérifier l'identité ou l'autorisation d'un utilisateur du système
61.
Hierarchical Data Modelling for Key-Value In-Memory Databases
A network component may include a memory and a processor. The processor may be configured to execute an in-memory data storage service configured to store data associated with the network component in key-value pairs within the memory, access a hierarchical schema mapping a hierarchical data structure to the key-value pairs within the memory, receive a command having a reference to the hierarchical data structure, identify an associated key-value pair associated with the reference to the hierarchical data structure, and implement the command by performing at least one of: an access action or a modifying action.
An information processing system includes a chassis having a front panel, a rear panel, and a pair of side panels. A module cage is received in and supported by the chassis. The module cage comprises a housing having a first bay and a second bay, each configured to removably receive a pluggable module. The module cage further comprises a duct assembly disposed in the housing and configured to define first and second airflow passages from outside of the module cage to the first and second bays, respectively. The duct assembly is passively actuatable using an actuatable flap from a first state to a second state in response to installation of a pluggable module in the second bay. In the first state, the duct assembly communicably connects the first and second airflow passages, while in the second state, the duct assembly separates the first and second airflow passages from one another.
Systems and methods are provided for controlling a plurality optical resonators along a common control bus. Examples include a plurality of resonators coupled to a plurality of tuning mechanisms, a control bus connected to the plurality of tuning mechanisms, and a resonator controller configured to input a data stream onto the control bus. The data stream includes a plurality of request data packets which are being assigned to respective ones of the plurality of resonators. The plurality of tuning mechanisms are configured to adjust resonance wavelengths of the plurality of resonators based on receiving the data stream.
G02B 6/293 - Moyens de couplage optique ayant des bus de données, c.-à-d. plusieurs guides d'ondes interconnectés et assurant un système bidirectionnel par nature en mélangeant et divisant les signaux avec des moyens de sélection de la longueur d'onde
A management controller of a computer platform manages a host of the computer platform. The management includes communicating, by the management controller and using a sideband channel interface of a network interface controller adapter, network traffic with a remote management service. The network interface controller adapter includes a host interface. The computer platform is configured to assign the network interface controller adapter to the management controller. Assigning the network interface controller to the management controller includes preventing the host from using the host interface.
G06F 21/85 - Protection des dispositifs de saisie, d’affichage de données ou d’interconnexion dispositifs d’interconnexion, p. ex. les dispositifs connectés à un bus ou les dispositifs en ligne
65.
HUMAN-INSPIRED MEMORY MANAGEMENT AND RESPONSE GENERATION
Systems and methods are provided for emulating human memory in the form of short-term memory (STM) and long-term memory (LTM) stores to which aspects of a chatbot conversation may be stored, and from which ST and LT memories may be retrieved based on relevance to the chatbot conversation and recency of access of ST memories. Contextual information relevant to the chatbot conversation may also be retrieved. A response to a query or input presented during the chatbot conversation can be generated using the retrieved ST/LT memories, the contextual information, and emotional tone of the chatbot conversation. The chatbot conversation itself can be stored as a memory(ies), and weights of memories, e.g., ST memories, can be adjusted to emulate human forgetfulness and remembrance.
H04L 51/02 - Messagerie d'utilisateur à utilisateur dans des réseaux à commutation de paquets, transmise selon des protocoles de stockage et de retransmission ou en temps réel, p. ex. courriel en utilisant des réactions automatiques ou la délégation par l’utilisateur, p. ex. des réponses automatiques ou des messages générés par un agent conversationnel
G06F 40/284 - Analyse lexicale, p. ex. segmentation en unités ou cooccurrence
H04L 51/04 - Messagerie en temps réel ou quasi en temps réel, p. ex. messagerie instantanée [IM]
An information processing system comprises a system board having electronic components and a chassis supporting the system board. The chassis includes fan shroud attachment points configured to receive a fan shroud attached thereto. A modular adapter for immersion cooling includes a support having a base and a pair of side arms. The side arms are configured to couple to the chassis at the fan shroud attachment points in lieu of the fan shroud. A manifold is coupled to the base and a pump is operatively coupled to the manifold. The pump is configured to move a flow of liquid coolant from a pool through the manifold when the information processing system is immersed in a pool of liquid coolant. The manifold is configured to output the flow of the liquid coolant back into the pool as one or more output streams.
An outdoor information processing device comprises a housing. The housing comprises a hollow interior volume sealed air-tight from an exterior environment and heat dissipation structures on an exterior of the housing. A coolant is disposed in the hollow interior volume of the housing. A portion of the coolant is in a liquid phase collected in a pool in the housing and a portion of the coolant being in a vapor phase above the pool. An information processing system board comprising electronic components is disposed in the hollow interior volume and immersed in the pool of the liquid phase of the coolant. The coolant is configured to transfer heat from the electronic components into the housing via the coolant going through a repeated phase-change cycle.
In some examples, a system identifies a memory page containing encrypted operating system (OS) information that is to be subjected to integrity monitoring, the encrypted OS information encrypted with a first key. The system associates an encryption disabled indicator with the memory page. The system requests the encrypted OS information from the memory page, wherein the request is to cause a central processing unit (CPU) to read the encrypted OS information from the memory page, and the encryption disabled indicator is to cause the CPU to refrain from decrypting the encrypted OS information using the first key when reading the encrypted OS information from the memory page. The system computes a reference measurement value based on the encrypted OS information read by the CPU from the memory page without decryption using the first key, the reference measurement value for use in the integrity monitoring.
G06F 12/14 - Protection contre l'utilisation non autorisée de mémoire
G06F 13/28 - Gestion de demandes d'interconnexion ou de transfert pour l'accès au bus d'entrée/sortie utilisant le transfert par rafale, p. ex. acces direct à la mémoire, vol de cycle
An axial pump for delivering liquid coolant to cool an electronic device comprises a conduit, an impeller in the conduit, a shaft, a rotor, and a stator. The conduit defines a flow path from an inlet of the conduit to an outlet of the conduit. The rotor is disposed in a hollow interior of the impeller and is rotatably coupled to the shaft. The stator is configured to drive rotation of the rotor about the shaft, with the impeller rotating along with the rotor about an axis of rotation extending parallel to the flow path. The stator comprises teeth that are heterogenous in shape, resulting in an asymmetrical stator, and reducing magnetic sticking of the rotor when powered off. A set of first stator teeth has a first profile shape and a set of second stator teeth have a second profile shape.
The present disclosure provides techniques for reimaging physical hosts from a first hypervisor to a second hypervisor. A method includes designating one of a plurality of physical hosts as a coordinating host, receiving a service request to reimage target hosts, obtaining installation media for the second hypervisor, and installing the second hypervisor on each target host. The installation is performed by streaming the installation media from the coordinating host to each of the target hosts. The method enables efficient transition between different hypervisors across multiple physical hosts in a coordinated manner.
G06F 9/455 - ÉmulationInterprétationSimulation de logiciel, p. ex. virtualisation ou émulation des moteurs d’exécution d’applications ou de systèmes d’exploitation
An example intrusion detection system and a computing system including the intrusion detection system are presented. The intrusion detection system includes a sensor, a trigger circuit, and a non-volatile memory (NVM) device. The sensor is configured to set a tamper detection signal in an active state responsive to detecting physical tampering. Further, the trigger circuit generates a status capture pulse when the tamper detection signal is in the active state. Furthermore, the NVM device receives the status capture pulse from the trigger circuit and the tamper detection signal from the sensor. The NVM device sets an intrusion detection signal to an active state at an output of the NVM device in response to receiving the status capture pulse and detecting that the tamper detection signal is in the active state, and maintains the intrusion detection signal in the active state when a power supply to the NVM device is restored.
According to an implementation, a credential rotation system provides automated coordination of credential updates across interdependent computing devices. A central manager receives component registrations containing credential configurations and dependency information. The manager maintains a dependency graph tracking which components rely on service accounts from other components, generates new credentials according to defined policies, and orchestrates synchronized updates through component-specific agents. When rotation is triggered for a component's service account, the manager identifies dependent components using the dependency graph and coordinates credential updates to maintain operational continuity. Component-specific agents handle local credential rotation and updates while the manager ensures system-wide coordination through dependency tracking and automated notification.
Systems and methods are provided for co-optimized recommendation and evaluation of models, such as foundation models. Users may specify, in a natural language query, a request for (or requirements) regarding one or more models for application to a given use case. The query can be characterized along with models of a model repository to determine the suitability of models to satisfy the query. Benchmarks of a benchmark repository may also be analyzed to determine their suitability to be used to properly test any recommended models. The recommended models can then be tested on the recommended benchmarks. The characterization and analysis of models and benchmarks can reduce the compute and time costs associated with manual/conventional model and benchmark selection for use in a downstream application.
Systems, methods, and devices for securely adding multihomed CE devices are provided herein. A MAC address for a second networking device connected to a networking device is received at the networking device via an Ethernet segment. The networking device generates a secure-type Ethernet segment identifier (ESI) for the second networking device. The networking device transmits the secure-type ESI to a third networking device. The secure-type ESI indicates that a port to the second networking device by the third networking device is to be kept operationally down until the second networking device is authenticated. The networking device authenticates the second networking device and sends the secure-type ESI in a subsequent message to indicate that authentication has been completed for the second networking device by the networking device. The subsequent message instructs the third networking device to make the port operational.
Systems and methods are provided for comprehensive observability into a cluster of nodes of a container management system for proactive compliance driven root-cause detection within the cluster. Examples train a root-cause analyzer for a cluster of a container management system by applying historical cluster observability data to a set of machine learning (ML) models. Examples detect technological anomalous events in the container management system and identify compliance events at compute nodes of the cluster based on cluster observability data received from the cluster. Examples predict a root-cause of the one or more technological anomalous events based on the compliance events and prioritize remedial actions to rectify the root-cause based on the compliance events and the technological anomalous.
Systems and methods are provided for comprehensive observability into a cluster of nodes of a container management system for proactive compliance driven root-cause detection within the cluster. Examples train a root-cause analyzer for a cluster of computer nodes of a container management system by applying historical cluster data obtained from the cluster of compute nodes to a set of machine learning (ML) models and predict a root-cause event in the container management system based on cluster data received from the cluster of compute nodes. Examples generate one or more software code candidates configured to fix the predicted root-cause, determine that a software code candidate of the one or more software code candidates satisfy security compliance criteria, and deploy the software code candidate in the container management system based on the determination.
G06F 21/52 - Contrôle des utilisateurs, des programmes ou des dispositifs de préservation de l’intégrité des plates-formes, p. ex. des processeurs, des micrologiciels ou des systèmes d’exploitation au stade de l’exécution du programme, p. ex. intégrité de la pile, débordement de tampon ou prévention d'effacement involontaire de données
Systems and methods are provided for implement a perturbation technique of a subset of input time series data (e.g., a “motif”) provided to a machine learning model to help determine the effect of the motif on the inference of the model, absent any knowledge of the weights/biases corresponding with that particular motif in the input data. The replacement of the motif may be implemented to help describe the significance of a repeating pattern in the time series data with respect to its effect on the inference output.
A technique includes receiving application intelligence data representing network traffic flow signatures that are associated with respective applications. The technique includes associating the applications with an application group and generating a tag associated with the application group and including data representing rules to apply to recognize the network traffic flow signatures. The technique includes configuring a managed network device to identify network traffic flows associated with the applications. The configuration includes providing the tag to the managed network device to cause the managed network device to apply the rules to identify the network traffic flows.
Network interconnect-offloaded triggered operations are provided herein, in which triggered management is offloaded from local processing circuitry of compute blades to network interconnects used to intercouple the compute blades. A network interconnect may receive trigger condition data, confirm whether a condition is met, and initiate chained operations when the condition is met with a reduced number of PCIe crossings, causing improvements in overall system latencies.
In certain implementations, a system includes one or more processors and one or more non-transitory computer-readable storage media storing programming for execution by the one or more processors. The programming includes instructions to shut down one or more virtual machines that are managed by a first hypervisor. The first hypervisor is running on a first host, and the one or more virtual machines are able to read and store data in a first filesystem in a storage volume. The programming includes further instructions to unmount the first filesystem from the first host by removing the first filesystem from an active file hierarchy of the first hypervisor, and to attach the storage volume to a second host on which a second hypervisor is running. The programming includes further instructions to create a second filesystem in the storage volume and mount the storage volume with the second filesystem on the second host.
A technique includes selecting paths of an interconnection fabric of a data center based on respective metric values that are associated with the paths. Each path extends from an associated host of the data center to an associated fabric-attached resource of the data center. The technique includes providing, to a fabric manager of the data center, a data structure that includes data representing the paths and the associated respective metric values. The technique includes receiving, by the fabric manger, a resource sharing request. The resource sharing request is associated with given hosts of the data center. The technique includes, responsive to the resource sharing request and based on the data structure, allocating a given fabric-attached resource to be shared by the given hosts.
A system receives, by an access switch in a network fabric, a first flow indicating a destination internal to the network fabric. The system records the first flow and a first reverse flow in a flow table. The system marks the access switch as an owner for the first flow and for the first reverse flow, and a respective owner provides information associated with a respective flow based on Internet Protocol Flow Information Export (IPFIX). The system exports information associated with the flows. The system receives a second flow, whose source has previously been marked as an owner for the second flow and for a second reverse flow. The system records the second flow and reverse flow in the flow table, refrains from marking the access switch as the owner for the second flow and reverse flow, and refrains from exporting information associated with the second flow and reverse flow.
Systems and methods are provided for performing frequency conversion using RF metamaterials. An RF metasurface element includes first and second antennas on first and second surfaces, respectively, the first and second surfaces being opposite one another. A modulation signal input is configured to receive a modulation signal. The RF metasurface element is configured to receive, at the first antenna, a first signal having a first frequency. The RF metasurface element is further configured to transmit a second signal at a second, different frequency, from the second antenna, wherein the second frequency is dependent on the first frequency and a frequency of the modulation signal. A plurality of RF metasurface elements may be assembled into an array of elements.
H01Q 15/00 - Dispositifs pour la réflexion, la réfraction, la diffraction ou la polarisation des ondes rayonnées par une antenne, p. ex. dispositifs quasi optiques
H01Q 21/06 - Réseaux d'unités d'antennes, de même polarisation, excitées individuellement et espacées entre elles
84.
Systems and methods for ensuring redundancy between relay agents
A computer-implemented method for ensuring consistency between redundant DHCP peers may include establishing, by a first relay agent over a TCP connection, a peer relationship with a second relay agent. The method may also include synchronizing, by the first relay agent, subscriber information with the second relay agent for a subscriber group. The method may then include periodically calculating, by the first relay agent, a checksum for all subscribers in the subscriber group. Additionally, the method may include exchanging, by the first relay agent, the checksum from the first relay agent with a second checksum from the second relay agent via periodic TCP keepalive messages. Furthermore, the method may include detecting, by the first relay agent, a mismatch between the checksums. Finally, the method may include resolving an inconsistency in subscriber information in response to detecting the checksum mismatch. Various other methods, systems, and computer-readable media are also disclosed.
H04L 45/28 - Routage ou recherche de routes de paquets dans les réseaux de commutation de données en utilisant la reprise sur incident de routes
H04L 61/5014 - Adresses de protocole Internet [IP] en utilisant le protocole de configuration dynamique de l'hôte [DHCP] ou le protocole d'amorçage [BOOTP]
H04L 61/5076 - Mécanismes de mise à jour ou de notification, p. ex. DynDNS
H04L 67/145 - Interruption ou inactivation de sessions, p. ex. fin de session contrôlée par un événement en évitant la fin de session, p. ex. maintien en vie, battements de cœur, message de reprise ou réveil pour une session inactive ou interrompue
A first node communicates, via a high availability link, initialization information with a second node and thereby configures the first node as an active node for a first service redundancy group (SRG) and as a backup node for a second SRG. The first node also causes a first ingress floating IP address to be associated with an ingress link of the first node and a first egress floating IP address to be associated with an egress link of the first node, wherein the first ingress floating IP address is associated with the first SRG and the first egress floating IP address is associated with the first SRG. The first node communicates, via the high availability link, SRG state information with the second node and updates a data structure based on the SRG state information.
H04L 41/0659 - Gestion des fautes, des événements, des alarmes ou des notifications en utilisant la reprise sur incident de réseau en isolant ou en reconfigurant les entités défectueuses
Systems and methods are provided for implementing encryption of data-in-motion and/or otherwise stored data using a key server and a secure enclave of a Network Interface Card (NIC). The NIC acts as a passthrough between the client device and the shared infrastructure of the supercomputer system to help ensure data security in a massively scaled and distributed system. For example, in response to an enrollment process that stores a decrypted key in the secure enclave of a NIC, the NIC can receive a data packet from a client device. The NIC can transmit a key request to a key server that includes an encrypted key corresponding to the decrypted key. The key server can look up the previously stored private/public key pair to authenticate the NIC. The key server can provide private/public key pair to the NIC to allow the NIC to later encrypt data-in-motion.
H04L 9/32 - Dispositions pour les communications secrètes ou protégéesProtocoles réseaux de sécurité comprenant des moyens pour vérifier l'identité ou l'autorisation d'un utilisateur du système
A computer platform includes a host and a management controller. The management controller operates independently from the host. A hardware processor of the management controller executes a firmware management stack to manage the host. A bus interface controller of the management controller generates bus signals to access a non-volatile memory. A secure enclave of the management controller includes a bus controller interface recovery engine to, responsive to the bus interface controller exhibiting an unresponsive behavior, communicate with the bus interface controller to reset the bus interface controller.
G06F 11/07 - Réaction à l'apparition d'un défaut, p. ex. tolérance de certains défauts
G06F 21/85 - Protection des dispositifs de saisie, d’affichage de données ou d’interconnexion dispositifs d’interconnexion, p. ex. les dispositifs connectés à un bus ou les dispositifs en ligne
88.
AUTO-MERGING DECENTRALIZED-TRAINING METADATA IN MACHINE LEARNING TRACKING PIPELINE
Systems and methods are provide for tracking metadata generated by distributed data processing pipelines. An empty execution template representative of a metadata tracking hierarchy can be distributed to a plurality of nodes comprising a distributed artificial intelligence (AI) network. Individual instances of the empty execution template which individual nodes of the plurality of nodes have populated with metadata arising out of the execution of one or more stages of an AI pipeline at the individual nodes, can be received from the plurality of nodes. The metadata from the individual nodes can be merged in accordance with a common execution identifier set forth in the execution template. A global AI pipeline with the merged metadata to facilitate distributed data processing operations with the merged metadata can be created.
In some examples, a system receives, at a controller, information from a client device as part of a registration process by the client device with the controller, the information received from the client device including switch information of a switch to which the client device is connected, the switch being part of a network fabric that connects client devices to destination devices. Based on the switch information, the system identifies the switch from among a plurality of switches of the network fabric as a target to configure an access control rule for a zone including the client device. The system configures, at the switch, the access control rule for the zone.
H04L 41/12 - Découverte ou gestion des topologies de réseau
H04L 67/1097 - Protocoles dans lesquels une application est distribuée parmi les nœuds du réseau pour le stockage distribué de données dans des réseaux, p. ex. dispositions de transport pour le système de fichiers réseau [NFS], réseaux de stockage [SAN] ou stockage en réseau [NAS]
90.
DYNAMICALLY INSTALLING AND UNINSTALLING GROUP-BASED POLICIES BASED ON APPLICATION OPERATING ENVIRONMENT ACTIVITY
A technique includes, on a network device, dynamically installing and uninstalling group-based policies based on application operating environment activity associated with the network device. Installing the group-based policies includes detecting a presence of an application operating environment on a network; responsive to detecting the presence of the application operating environment, determining Internet Protocol (IP) address of the application operating environment; determining a role that is associated with the application operating environment based on the IP address; identifying a given group-based policy corresponding to the role; and installing, by the group-based network device, the given group-based policy on the network device to add the given group-based policy to the composition of group-based policies installed on the network device. The network device regulates network traffic flows received by the network device to enforce the given group-based policy.
H04L 41/0894 - Gestion de la configuration du réseau basée sur des règles
H04L 45/745 - Recherche de table d'adressesFiltrage d'adresses
H04L 61/103 - Correspondance entre adresses de types différents à travers les couches réseau, p. ex. résolution d’adresse de la couche réseau dans la couche physique ou protocole de résolution d'adresse [ARP]
91.
Data validation of container objects in a deduplication storage system
Example implementations relate to operations in a storage system. An example includes receiving a data unit to be stored in a persistent storage of a deduplication storage system, and identifying a container index that includes metadata regarding the data unit. The example also includes identifying a first container entity group (CEG) object that is recorded, in the container index, as a deduplicated storage location of the data unit. The example also includes, in response to a determination that that a validation state of the first CEG object is recorded as invalid in the container index, storing the data unit in a second CEG object. The example also includes modifying the container index to record the second CEG object as the deduplicated storage location of the data unit.
A device may receive a request for service requirements, and may determine network devices that satisfy the service requirements. The device may identify an unreachable network device from the network devices determined to satisfy the service requirements, and may deny the request based on the unreachable network device being unreachable. The device may receive a command to proceed with the request, and may create an orphan service instance for resources used by the unreachable network device and service instances for resources used by the remaining network devices. The device may associate a first device status with the orphan service instance, and may associate second device statuses with the service instances. The device may generate a service design based on the orphan service instance, the service instances, the first device status, and the second device statuses, and may utilize the service design to satisfy the service requirements.
H04L 41/5054 - Déploiement automatique des services déclenchés par le gestionnaire de service, p. ex. la mise en œuvre du service par configuration automatique des composants réseau
H04L 41/0668 - Gestion des fautes, des événements, des alarmes ou des notifications en utilisant la reprise sur incident de réseau par sélection dynamique des éléments du réseau de récupération, p. ex. le remplacement par l’élément le plus approprié après une défaillance
H04L 43/0811 - Surveillance ou test en fonction de métriques spécifiques, p. ex. la qualité du service [QoS], la consommation d’énergie ou les paramètres environnementaux en vérifiant la disponibilité en vérifiant la connectivité
93.
TRACKING APPLICATION BEHAVIOR BASED ON PACKET SPACING
A system tracks, in a network device with multiple links, activity over a respective link for a predetermined amount of time. The activity comprises at least one of a number of idle periods or a duration of a respective idle period. The system divides a predetermined time interval into a number of bins. A bin is associated with a range of time. The system stores the tracked activity in data structure entries which each indicate the number of bins, a duration of time associated with a respective bin, and a count associated with the respective bin. The system indicates the tracked activity in the entry for the respective link by incrementing a count associated with a bin matching the duration of the respective idle period based on a first number of idle periods tracked for the matching duration. The system displays the stored tracked activity for the respective link.
This disclosure is directed to a network switch with decentralized processing threads for accessing a shared database. The network switch may include database may include tables storing data entries, and processing circuitry that may define independent processing threads. Each processing thread may implement a packet processing pipeline including a shared database logic of the database. The packet processing pipelines may access data entries by referencing the database logic to perform network operations. Each packet processing pipeline may use semi-lockless atomic operations with temporary spin-locks to access target data entries without locking access of other packet processing pipelines to other data entries. As such, the packet processing pipelines may access an increased number of data entries of the database simultaneously. Moreover, the processing circuitry may allocate portions of the database to different packet processing pipelines to reduce contention and waiting times when accessing the database using the semi-lockless atomic operations.
G06F 16/27 - Réplication, distribution ou synchronisation de données entre bases de données ou dans un système de bases de données distribuéesArchitectures de systèmes de bases de données distribuées à cet effet
A system for exchanging data between a baseboard management controller (BMC) and a host processor is described herein. The system includes one or more host processors and a bus interface. The system also includes BMC circuitry that includes a BMC microcontroller configured to perform instructions received from one or more remote electronic devices to enable the one or more remote electronic devices to manage a server device on which the BMC circuitry resides. The system further includes a shared memory that is accessible by the one or more host processors via the bus interface and the BMC microcontroller. The one or more host processors and the BMC microcontroller are configured to exchange data with each other via the shared memory.
In some examples, a first network device receives, from a second network device, role update information for a compute entity connected to the second network device, the role update information including role information specifying a role of the compute entity and a network address of the compute entity. The first network device updates role mapping information stored at the first network device based on the received role update information, where the role mapping information correlates network addresses to roles. The first network device receives, from a first compute entity, a data packet containing a destination network address of a second compute entity to which the data packet is targeted. The first network device performs a lookup of the role mapping information using the destination network address to determine a role of the second compute entity, and enforces a policy with respect to the data packet at the first network device based on a role of the first compute entity and the role of the second compute entity.
Example implementations relate to deduplication operations in a storage system. An example includes detecting metadata changes associated with a container index of the deduplication storage system, and recording the metadata changes in multiple journals included in a journal group. The example also includes identifying, in the journal group, a first journal having a filled amount that is less than a fold threshold. The example also includes, determining whether the identified first journal is stagnant based on one or more journal metrics of the identified first journal, and in response to a determination that the identified first journal is stagnant based on the one or more journal metrics, modifying the container index to include each metadata change recorded in the identified first journal.
In some examples, a system receives a collection of values of a plurality of metrics including a resource utilization metric representing utilization of a resource, and a data communication metric representing communication of data. The system detects an anomalous behavior of a first metric of the plurality of metrics. Based on detecting the anomalous behavior of the first metric, the system computes a measure of correlation between the first metric and at least a second metric of the plurality of metrics. The system determines whether an unauthorized data transfer is occurring based on the measure of correlation.
In an overlay network, a network device can receive a first notification message indicating that a host coupled to a second network device of the overlay network has requested to join a multicast group. The network device can generate a mapping between the host and the multicast group based on the first notification message. Upon detecting the host via a port, the network device can generate a second notification message based on the mapping prior to receiving a join request from the host. The network device can send the second notification message to a respective other network device of the overlay network. The network device can receive a set of multicast packets of the multicast group via a tunnel coupled to a source network device in response to the source network device receiving the second notification message and forward the set of multicast packets via the port.
A network device may receive parameters for a probe packet, and may generate the probe packet based on the parameters. The network device may provide the probe packet to multiple ingress packet forwarding components, and may generate probe packet copies at the multiple ingress packet forwarding components. The network device may generate ingress route metadata for the probe packet copies via filters provided in the multiple ingress packet forwarding components, and may provide the probe packet copies to multiple egress packet forwarding components. The network device may generate egress route metadata for the probe packet copies via filters provided in the multiple egress packet forwarding components, and may provide the probe packet copies, the ingress route metadata, and the egress route metadata to an application. The network device may utilize the application to generate a report, and may provide the report for analysis.
