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.
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.
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
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.
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.
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.
Techniques are disclosed for mitigating lost state data for a communication session by a network device performing session-based routing. For example, in response to determining a first network device has not received packets of a reverse packet flow of a session from a second network device, the first network device sends, to the second network device, a first packet including first metadata comprising a first control message querying a health of the session. The first network device receives, from the second network device, a second control message requesting second metadata for reestablishing the session, the second control message sent in response to the first control message. The first network device sends a second packet of the forward packet flow including the second metadata for reestablishing the session to the second network device. The second network device recovers lost state data for the session using the second metadata.
H04L 12/28 - Réseaux de données à commutation caractérisés par la configuration des liaisons, p. ex. réseaux locaux [LAN Local Area Networks] ou réseaux étendus [WAN Wide Area Networks]
H04L 43/106 - Surveillance active, p. ex. battement de cœur, utilitaire Ping ou trace-route en utilisant des informations liées au temps dans des paquets, p. ex. en ajoutant des horodatages
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
H04L 67/146 - Marqueurs pour l'identification sans ambiguïté d'une session particulière, p. ex. mouchard de session ou encodage d'URL
10.
Restoring container connectivity for containerized routers
In general, this disclosure describes techniques for automatically re-attaching containers to a containerized router after uninstallation and re-installation of the containerized router. A containerized router may read container attachment information from memory, where the container attachment information is persisted to memory by a prior instance of the containerized router. The containerized router may configure the containerized router with the container attachment information to enable network communications by a logically-related group of one or more containers via the containerized router.
A server includes physical compute nodes. Each physical compute node includes a host and physical management resources. The physical management resources include a physical management processor. The server includes a distributed hypervisor to provide a distributed application operating environment that is hosted by the physical management resources. The distributed hypervisor to allocate, from the physical management processors, virtual processors for the distributed application operating environment to execute applications to manage the physical compute nodes. The distributed hypervisor includes a plurality of hyper-kernels that are associated with respective physical compute nodes. Each hyper-kernel is hosted on the physical management resources of the associated physical compute node.
G06F 9/455 - ÉmulationInterprétationSimulation de logiciel, p. ex. virtualisation ou émulation des moteurs d’exécution d’applications ou de systèmes d’exploitation
12.
CLASSIFICATION OF NETWORK TRAFFIC USING A CENTRALIZED CACHE
In certain implementations, a computing device includes a centralized cache storing classification information for network traffic associated with access points (APs), 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 receive a classification request from a first AP for network traffic associated with a client device, and in response, determine whether the centralized cache includes classification information for the network traffic. The programming includes instructions to obtain, in response to determining that the centralized cache includes the classification information, the classification information from the centralized cache. The programming includes instructions to obtain, in response to determining that the centralized cache does not include the classification information, the classification information from a traffic classification service. The programming includes instructions to transmit the classification information from the traffic classification service to the first AP.
Load balancing operations between network devices to perform DPI operations are provided herein. Network devices of a virtualized network device arrangement may load balance network traffic by employing a network link between the network devices to exchange data packets for DPI operations. Further, the network devices may employ dedicated CPUs to offload DPI operations. In this manner, the current techniques may enable preservation of pre-existing networking routing or deterministic routing to be performed on-the-fly by enabling DPI to be performed without constraining network traffic flows to a specified route.
Systems and methods are provided for a multi-tenant collective communication fabric for optimal utilization of communication links between compute nodes of an interconnected system. Examples include allocating a plurality of compute nodes to a first workload and obtaining a topology of the interconnected system representative of an indirect path between the allocated compute nodes. The indirect path comprises a non-allocated compute node of the interconnected system. The examples include creating plurality of slices of resources of the non-allocated compute node, with a first slice dedicated to processing and forwarding data traffic along the indirect path and a second slice configured for allocation to a second workload. The examples also include executing the first workload by the allocated compute nodes and the non-allocated compute node, wherein data traffic from the plurality of allocated compute nodes is communicated via the indirect communication path.
A method for optimizing a topology of a mesh network. The method comprises determining a topology of a mesh network and a first access point (AP) to be added to the mesh network. The method further comprises determining a multi-uplink group between the first AP and the second AP. The method further comprises determining a first number of sibling APs of the first AP and a second number of neighbor APs of the first AP by determining the second AP as a parent AP of the first AP. The method further comprises determining a metric value for the multi-uplink group based on the first number of sibling APs and the second number of neighbor APs of the first AP. The method further comprises adding, based on determining that the metric value meets a predetermined condition, the first AP into the mesh network according to the multi-uplink group.
A system and method for converting and copying a virtual disk image between provider-specific computing resources is provided. The method includes receiving a request to copy a virtual disk image from a source to a destination, and converting the virtual disk image from a source format to a destination format while copying. The conversion process involves predicting locations of data blocks within the virtual disk image based on structural characteristics of the source format without accessing metadata at the end of the image. Data blocks are decoded from the source format to a raw format while streaming from the source, based on the predicted locations. The data blocks are then encoded from the raw format to the destination format while streaming to the destination. This method enables efficient conversion and transfer of virtual disk images between different provider-specific computing resources.
A fan interconnect module for an information processing device comprises a housing. A fan connection connector is formed in the housing and may be configured to selectively receive an electrical connector of a fan module. A cable connection connector is further formed in the housing and be configured to selectively receive a first cable connector of a cable connected to a system board of the information processing device. The cable connection connector may be electrically connected to the fan connection connector and may have an insertion axis perpendicular to the insertion axis of the fan connection connector. The fan interconnect module may include first attachment features on a first side of the housing and configured to mount the fan interconnect module in a first orientation and second attachment features on a second side of the housing and configured to mount the fan interconnect module in a second orientation.
includes a movable member, a first power input connector, and a second power input connector. The movable member is configured to be movably connected to a chassis of the information processing device such that the movable member is movable outside the chassis to an extended position and inside the chassis to a retracted position. The first and second power input connectors are disposed serially spaced apart from each other along an axis parallel to a direction of movement of the movable member and coupled to the chassis via the movable member. The movable member supports the first and second power input connectors relative to the chassis. The first and second power input connectors are configured to travel along with the movable member relative to the chassis as the movable member moves between the extended position and the retracted position.
H05K 7/14 - Montage de la structure de support dans l'enveloppe, sur cadre ou sur bâti
H01R 13/518 - Moyens pour maintenir ou envelopper un corps isolant, p. ex. boîtier pour maintenir ou envelopper plusieurs pièces de couplage, p. ex. châssis
Systems and methods are provided for optimizing, e.g., speeding up, fine time measurement (FTM) scanning by dedicating a radio of a network element, such as an access point (AP), to performing FTM scanning at a high frequency. The high frequency FTM scanning provides quicker results regarding location determination of the network element as compared to conventional FTM scanning, which is performed merely as a background process. The high frequency FTM scanning can be performed in accordance with configurable FTM scan parameters.
H04B 7/06 - Systèmes de diversitéSystèmes à plusieurs antennes, c.-à-d. émission ou réception utilisant plusieurs antennes utilisant plusieurs antennes indépendantes espacées à la station d'émission
H04B 7/212 - Accès multiple par répartition dans le temps
H04B 17/18 - Surveillance en fonctionnement normal
In implementations of the present disclosure, there is provided an approach for an enhancement mechanism for fine time measurement. A method comprises detecting, by an access point (AP) multi-link device (MLD), a set of parameters for each of a plurality of links between the AP MLD and a set of station MLDs. Then, the AP MLD may use the set of parameters to determine a score for each of the plurality of links. The plurality of scores for the plurality of links may be used to select a set of links from the plurality of links for transmitting a plurality of FTM frames. The AP MLD further determines a number of the FTM frames to be transmitted on one of the set of links. Then, the AP MLD transmits the number of the FTM frames to the set of station MLDs via the one of the set of links.
Systems and methods for managing communications between provider-specific computing resources and a central management platform are provided. A persistent connection is established between a management worker and the central management platform, which may be deployed in a first network. The management worker receives connection requests from agents of provider-specific computing resources, which may be deployed in a second network. A multiplexed message stream is established using the persistent connection in response to the connection requests. The management worker relays messages between the agents and the central management platform through the multiplexed message stream, enabling efficient communication across different networks.
An information processing system includes a chassis and a system board supported by the chassis and having a processor. The chassis further supports a plurality of fans. A cage having a front bracket and a pair of side brackets is coupled to either the system board or the chassis, with the cage defining at least two bays to receive pluggable modules. The front bracket further includes at least one pair of cable management supports extending forward from the front bracket. At least two cables each comprising a first cable connector are coupled to the front bracket and arranged to mate with the pluggable modules. Each pair of the cable management supports is arranged to support a flexible wire portion of one of the cables at a location forward of the front bracket and hold the flexible wire portion at a position higher than an airflow opening in the front bracket.
Systems and methods are provided for failure resiliency in distributed training of machine learning (ML) models. Examples include a plurality of compute nodes storing shards of a plurality of shards of model states of an ML model, and a first compute node storing a first shard of model states of the ML model. The first compute node can store a plurality of shard portions. Each shard portion can be received from a respective compute node of the plurality of compute nodes and can be a replica of a portion of a respective shard, of the plurality of shards, stored at the respective compute node. Responsive to a failure of a compute node of the plurality of compute nodes, the first compute node can update the first shard with a shard portion corresponding to the failed compute node and the ML model can be trained based on the updated first shard.
G06F 11/20 - Détection ou correction d'erreur dans une donnée par redondance dans le matériel en utilisant un masquage actif du défaut, p. ex. en déconnectant les éléments défaillants ou en insérant des éléments de rechange
G06N 3/098 - Apprentissage distribué, p. ex. apprentissage fédéré
24.
Optimization of radio resource management (RRM) based on channel assignments
In an example, a method includes, for each of a plurality of access points (APs) in a wireless network, automatically configuring the AP with a channel-specific transmit power optimization corresponding to an assigned channel of a plurality of channels of a given frequency band, wherein the channel-specific transmit power optimization is based on a difference between attenuation of wireless signals transmitted using the assigned channel of the plurality of channels and attenuation of wireless signals transmitted using a baseline channel of the plurality of channels.
Examples described herein relate to allocating a low latency, healthy fabric-attached memory resource to a host device for downloading media. A resource composer may receive, from a host device, a request to download media from a pool of fabric-attached memory resources. The resource composer may identify, from the pool, a set of memory resources based on resource latency. The resource composer may group, based on a respective error history of a given memory resource, the set into a first group and a second group, where the respective error history of each of the memory resources in the first group includes fewer errors than the second group. The resource composer may select, from the first group, a memory resource to allocate to the host device and may allocate the memory resource to the host device. The media may be accessible to the host device at the allocated memory resource.
A voltage regulator module and a circuit board of an information processing device can be removably connected together via a reversible mechanical connection. The voltage regulator module comprises a printed circuit board (PCB), voltage regulator components mounted to one face of the PCB, and electrical contacts arranged as concentric rings in an opposite face of the PCB and electrically connected to the voltage regulator components through the PCB. The module also comprises at least one attachment feature configured to removably attach the module to the circuit board. The concentric rings are arranged to contact solder balls arranged in a complementary pattern on a face of the circuit board, thereby electrically connecting the module to the circuit board. The contact between the rings and the solder balls may be simple mechanical contact without any soldered (i.e., braised or welded) bond formed between the solder balls and the rings.
An information processing system comprises a rack capable of accommodating servers and a power distribution circuitry that provides power to the servers. The system further includes a power input module (PIM) placed in the rack that receives input power and converts the input power to output power. The PIM conveys the output power to the power distribution circuitry. The PIM includes power circuitry that converts the input power to the output power, and a housing that houses the power circuitry. The housing has a fixed portion, an input assembly that routes input feeds from outside of the rack into the PIM, and an output assembly comprising output feeds that supply the output power to the power distribution circuitry. The input assembly and the output assembly are movable relative to the fixed portion of the housing.
Accelerator arrangement optimization operations are provided herein. A workload re-distribution system may estimate performance metrics associated with one or more accelerator orchestrations and select, based on the estimated performance metrics, an optimized accelerator arrangement to implement. An application workload may be processed based on the implemented optimized accelerator arrangement. In this manner, the overall efficiency of a computing infrastructure may improve by reducing computing resources spent implementing unoptimized accelerator orchestrations.
Techniques are provided for identifying and/or mitigating performance variability with respect to execution of application over a number of executions. Specifically, performance metrics are obtained for a number of executions of the application until a statistically stable distribution of the performance metrics is attained. Variability and execution factors corresponding to the variability and/or mitigation factors associated with execution factors may be identified and presented, enabling variability reduction for subsequent executions of the application.
Techniques are provided for reducing the amount of information to be transmitted to a user about resources of a cloud infrastructure, while allowing the information available to the user to remain current relative to the state of the resources in the cloud infrastructure. A cache ring is generated for a resource type, and the cache ring comprises a plurality of entries corresponding to objects associated with a resource/resources of the resource type in the cloud infrastructure. The plurality of entries includes a set of entries corresponding to a full refresh having a start index and an end index of the cache ring. When a subscription request is received from a user interface, a partial refresh is performed for the user interface based on the full refresh if certain conditions are satisfied, otherwise, another full refresh is performed for the user interface.
G06F 12/0891 - Adressage d’un niveau de mémoire dans lequel l’accès aux données ou aux blocs de données désirés nécessite des moyens d’adressage associatif, p. ex. mémoires cache utilisant des moyens d’effacement, d’invalidation ou de réinitialisation
31.
PREDICTIVE COMPONENT MANAGEMENT WITH AUTOMATED REPLACEMENT AND DISPOSAL DETERMINATIONS
Systems and methods are provided for monitoring operations performed by a device component to help implement predictive component management with automated replacement and disposal determinations for the device component. For example, the system may monitor operations performed by the component through a first time (e.g., using sensors, operational logs, etc.). The system may inspect the physical parameters of the device component, with analytics from a machine learning model to help determine a predicted value of the device component. In response to the predicted value, the system may initiate an action associated with the predicted value.
In some examples, a system produces tags based on a plurality of tests run in a computing system including program components, where each tag of the tags indicates a relevance of a respective test to a corresponding program component in the computing system. The system receives an indication of at least one changed program component from among a plurality of program components executable in the computing system, and the system filters the plurality of tests based on the tags and the at least one changed program component to identify a subset of tests from among the plurality of tests. The system triggers performance of the subset of tests in the computing system.
Systems and methods are provided for adjusting runtime environments for optimizing performance. For example, the system can deploy a monitoring agent configured to collect a metrics of a runtime environment and provide them as input to a first machine learning model that outputs a pattern in the metric of the runtime environment. The output is provided to a second model that compares the pattern with a metric threshold. In response to the comparison, the system may determine an adjustment to the parameter that changes memory management or compilation strategies of the device. The system may also receive real-time feedback of an effect of the adjustment in the runtime environment and retrain the second model with the real-time feedback.
G06F 9/455 - ÉmulationInterprétationSimulation de logiciel, p. ex. virtualisation ou émulation des moteurs d’exécution d’applications ou de systèmes d’exploitation
G06F 9/50 - Allocation de ressources, p. ex. de l'unité centrale de traitement [UCT]
34.
Ping and traceroute in inter-autonomous system (AS) segment routing (SR) networks without requiring headend router or path monitoring system (PMS) controller knowledge of topology outside of origin AS
Ping or traceroute functionality is supported in a path spanning multiple autonomous systems (ASes) having segment routing (SR) enabled, the path including an ingress node in a first autonomous system (AS) and an egress node in an AS other than the first AS, using a reverse path label pair including (1) a node segment identifier (SID) corresponding to an AS Border Router (ASBR) of the second AS (second ASBR), and (2) an egress peer engineering (EPE) SID corresponding to a segment between the second ASBR to an ASBR of the first AS (first ASBR). Responsive to receiving a ping or traceroute request by a router in the second AS, the router generates a ping or traceroute reply including the reverse path label pair. The ping or traceroute reply is forwarded to the second ASBR using the node SID of the reverse path label pair. The ping or traceroute reply is then forwarded from the second ASBR to the first ASBR using the EPE SID of the reverse path label pair. Finally, the ping or traceroute reply can be forwarded (e.g., using standard IP forwarding) from the first ASBR to the headend router.
H04L 45/50 - Routage ou recherche de routes de paquets dans les réseaux de commutation de données utilisant l'échange d'étiquettes, p. ex. des commutateurs d'étiquette multi protocole [MPLS]
H04L 45/00 - Routage ou recherche de routes de paquets dans les réseaux de commutation de données
H04L 45/02 - Mise à jour ou découverte de topologie
35.
MULTI-TRAFFIC-CLASS TRACKER ARBITRATION WITH FOCUS AND PRIORITIZED DEALLOCATION
A direct memory access (DMA) engine receives an instruction corresponding to a packet of a message and indicating a traffic class identifier (ID). The engine stores an entry for the instruction in a tracker, a respective tracker entry comprising a traffic class ID and indicators of whether the respective entry requires processing, a transfer of a DMA payload is complete, and is actively transferring the DMA payload. The engine arbitrates among the entries, by: determining a current mask indicating entries currently excluded selection; updating the current mask in cycles in response to detecting remaining tracker entries which require processing, have sufficient available output queue credit for the traffic class ID, and are not currently masked; and identifying the winning tracker entry based on the current mask and the indicators for the entries. The system forwards information associated with the winning tracker entry in response to meeting a priority level.
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
In certain examples, a method includes obtaining, from a schema and by a disconnected private cloud update packager, an information set corresponding to a set of updates for a disconnected private cloud; obtaining, based on the information set and by the disconnected private cloud update packager, the set of updates from one or more update locations; packaging, by the disconnected private cloud update packager, the set of updates as a disconnected private cloud update package comprising the set of updates, wherein the disconnected private cloud update package, when deployed in the disconnected private cloud, updates a plurality of components of the disconnected private cloud; and providing the disconnected private cloud update package to one or more entities that maintain disconnected private cloud instances.
In some examples, a system computes, based on metric information of a computing environment, a lower bound and an upper bound of resource usage in the computing environment. Based on the lower bound and the upper bound, the system derives a plurality of candidate resource request values. For each respective candidate resource request value, the system computes a corresponding score representing a sufficiency of the respective candidate resource request value and resource waste associated with the respective candidate resource request value. Based on the scores, the system elects a resource request value from the plurality of candidate resource request values. The system provides the selected resource request value to a service for inclusion in a scheduling request to schedule a workload of the service in the computing environment.
A docking device, a system and a method. The device comprising a bracket assembly, a pivoting retainer pivotally attached to the bracket assembly and a translating holder located within the bracket assembly, wherein the pivoting retainer is configured pivotally move between a latched state and an unlatched state and the translating holder is configured to translate between a first and a second position, wherein the translating holder provide structural support to M.2 modules in the first position.
An example method and a network controller are presented to aid in time-synchronizing networking resources. The network controller may create a network domain comprising a set of candidate network resources based on a time-sensitive networking demand of the set of candidate network resources. Further, the network controller may select a reference clock resource for the network domain from a set of candidate networking resources based on respective time-synchronization capabilities. Furthermore, the network controller may determine target times corresponding to affiliate networking resources based on respective predefined network delays relative to the reference clock resource, wherein the affiliate networking resources are candidate networking resources of the network domain other than the reference clock resource. After the target times are calculated, the network controller may cause the reference clock resource to transmit the target times to the affiliate networking resources.
A computing system may include an optimization preprocessing circuit configured to represent a SAT problem including a plurality of variables and one or more clauses including variables; receive proposed input values for the variables of the SAT problem; output violation indication information for the one or more clauses of the SAT problem according to the proposed input values; and calculate transition cost values for the variables in the SAT problem. The computing system may include leaky integrate-and-fire (LIF) circuitry configured to implement a plurality of LIF neurons, wherein each LIF neuron corresponds to a transition cost value of the transition cost values calculated by the optimization preprocessing circuit; capture the transition cost values in the LIF neurons; evaluate the transition cost values to determine variable selection indicators for determining new proposed input values for the variables of the SAT problem; and output a result based on the variable selection indicators.
Memory device controlled is described herein. Specifically, a computing system includes a memory device and a processor. The processor is configured to operate the memory device for a first period of time using a first power state descriptor for the memory device. Moreover, the first power state descriptor is lower than a performance target for the memory device. The processor is further configured to operate the memory device within a threshold of the performance target by switching from the first power state descriptor to a second power descriptor for the memory device. Furthermore, the second power state descriptor is higher than the performance target for the memory device.
In some examples, an electronic module includes a nonvolatile memory storing a certificate chain of certificates, the certificates including an attribute certificate of the electronic module, where a public key is included in the attribute certificate. The electronic module includes a module controller to request a certificate from a processor in a compute platform in which the electronic module is placed, receive, from the processor in the compute platform, a signed version of the certificate as signed using a private key, and authenticate the compute platform by using the public key in the attribute certificate to decrypt the signed version of the certificate.
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
H04L 9/14 - Dispositions pour les communications secrètes ou protégéesProtocoles réseaux de sécurité utilisant plusieurs clés ou algorithmes
43.
SMART PERFORMANCE ANALYSIS AND ANOMALY DETECTION FOR HPC AND AI SYSTEMS
Techniques for analyzing performance and detecting anomalies for complex large-scale systems are provided herein. More specifically, the present disclosure provides the ability to identify events of interest in counter samples using markers and correlating metrics from different entities (switch, NIC, CPU, GPU, PCIe, memory). In contrast to previous visual representations, the correlated events of the current techniques enable an enhanced visual representation that focuses on the identified events and a timeline correlation of different metrics to understand job performance variations.
G06F 3/04815 - Interaction s’effectuant dans un environnement basé sur des métaphores ou des objets avec un affichage tridimensionnel, p. ex. modification du point de vue de l’utilisateur par rapport à l’environnement ou l’objet
G06T 1/20 - Architectures de processeursConfiguration de processeurs p. ex. configuration en pipeline
44.
ACCELERATED COMPUTATION OF DIRECT MEMORY ACCESS SCATTER CONTEXT FOR GET RESPONSE
A system receives an instruction corresponding to a Get request packet of a message and indicating a pattern type associated with direct memory access (DMA) write operations for the Get response. The system determines a descriptor and starting context associated with the Get request packet if the type of pattern indicates nested loops associated with a multi-dimensional array structure. The system stores the starting context in a hardware table, providing access to the starting context in response to processing a Get response packet corresponding to the Get request packet. The system processes the instruction in cycles until a byte count of bytes hypothetically transferred is equal to or greater than a size of the Get request payload. The system obtains an ending context comprising updated loop counters and byte offset and stores the ending context in a cache as the starting context for a next instruction of a same message.
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 15/80 - Architectures de calculateurs universels à programmes enregistrés comprenant un ensemble d'unités de traitement à commande commune, p. ex. plusieurs processeurs de données à instruction unique
In certain examples, a method includes receiving, from a requesting entity and at an authentication and authorization adapter, an incoming request token, an app token, and a requested action set; assessing, by the authentication and authorization adapter, the incoming request token to determine an incoming request token type; providing, by the authentication and authorization adapter, the incoming request token and the requested action set to a particular identity access management (IAM) system of a plurality of IAM systems of a private cloud based on the incoming request token type; and providing, by the authentication and authorization adapter, an access response to the requesting entity based on an IAM response from the particular IAM system.
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
In a network, a network device can receive, at the forwarding hardware of the network device from a processor of the network device, a packet with an indicator indicating the packet to be a delay measurement packet via a processor port of the processor. The network device can tag, using the forwarding hardware, the packet with a first timestamp indicating a time when the forwarding hardware receives the packet. The network device can then forward, using the forwarding hardware, the packet to an egress queue of an egress network port of the network device. The network device can retrieve, using the forwarding hardware, the packet from the egress queue and tag the packet with a second timestamp indicating a time when the forwarding hardware retrieves the packet for forwarding. The network device can include, using the forwarding hardware, a delay indicated by the first and second timestamps into the packet.
H04L 47/283 - Commande de fluxCommande de la congestion par rapport à des considérations temporelles en réponse à des retards de traitement, p. ex. causés par une gigue ou un temps d'aller-retour [RTT]
47.
ROLE DETERMINATION IN A NETWORK ADDRESS ASSIGNMENT PROCESS
In some examples, a server provides a network address assignment service for compute entities to assign network addresses to the compute entities. The server receives a first message including an indicator for a compute entity, the first message being part of a network address assignment process for the compute entity, and the indicator informing the server that the server is to assign a role to the compute entity for implementing a role-based policy. The server determines, based on detecting the indicator, the role of the compute entity. The server sends, as a response to the first message, a second message containing a role field specifying the role of the compute entity.
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]
In some examples, a first electronic device includes an interface to communicate over a plurality of virtual networks. A memory stores a mapping between virtual stream identifiers and respective interface-virtual network combinations, where a virtual stream identifier identifies a communication stream over a virtual network. As part of a time synchronization process over the virtual network, a first time clock perform a lookup of the mapping using a first virtual stream identifier to retrieve an identifier of the virtual network. The first time clock adds the identifier of the virtual network to a timing message, and sends, over the virtual network, the timing message to a second electronic device that includes a second time clock to which the first time clock of the first electronic device is to be synchronized over the virtual network according to the time synchronization process.
Systems and methods are provided for predicting memory capacity issues at a user device by analyzing sparse and/or sporadic alert data generated by the user device. Using the time series dataset, the system can detect an anomaly based on the alert data. The system can initiate an action associated with the detected anomaly (e.g., regarding memory usage, adding memory, reclaiming memory using a local command, and other actions).
A network management system (NMS) automatically determines floor information for group(s) of one or more deployed access points (APs) in a wireless network. The floor information may include, for example, a specific floor number of a multi-floor structure on which each of the group(s) of APs is installed.
In some implementations, a key performance indicator (KPI) monitoring controller device may receive, from a user interface device, an indication that a first key-set topic associated with a network device is to be monitored. The KPI monitoring controller device may determine a first key-set threshold, which may be a quantity of key-set instances that are permitted to be monitored for the network device, and which is selected from multiple candidate key-set thresholds based on a subscription status associated with the network device. The KPI monitoring controller device may determine a first set of key-set instances to be monitored by a KPI monitoring collector device, with a quantity of key-set instances associated with the first set of key-set instances being less than or equal to the first key-set threshold. The KPI monitoring controller device may transmit, to the KPI monitoring collector device, an indication of the first set of key-set instances.
H04L 43/045 - Traitement des données de surveillance capturées, p. ex. pour la génération de fichiers journaux pour la visualisation graphique des données de surveillance
In some examples, a first network device performs an election process with other network devices to determine which of a plurality of network devices is to be a provider of a first program image module of a program image divided into multiple program image modules. As part of the election process, the first network device exchanges information of operational properties of the plurality of network devices. Based on the information of operational properties of the plurality of network devices, first network device determines whether the first network device is elected as the provider of the first program image module.
Examples disclosed herein relate to providing unified configuration of heterogeneous devices in a network. A request in a unified configuration format to configure a parameter of a device in a network may be received at a system. The system identifies a network device targeted by the request. Based on a command mapping, one or more commands usable to change a setting of the parameter at the target network device may be determined, and based on a command hierarchy, an order for the one or more commands may be determined. The system arranges the commands in the determined order to produce a command block having a device-specific configuration format for the target network device and causes the parameter to be set at the target network device. Based on the request, the system may further produce a command block having a second device-specific configuration format for a second device.
Systems and methods are provided for a waveband architecture in which waveband spacing can be adapted to accommodate varying numbers of wavelength channels per waveband. Examples include an optical source that generates an input light beam. The input light beam includes sets of optical signals within wavebands and the sets of optical signals include a plurality of optical signals at wavelengths within respective wavebands. Examples also include waveguides that receive the sets of optical signals, which are split amongst the plurality of waveguides according to the wavebands. Examples include sets of optical modulators coupled to the waveguides, where each optical modulator modulates an optical signal at a resonance wavelength of the respective optical modulator. The wavebands comprise a waveband spacing that is based on a free spectral range of the optical modulators. A multiplexer combines the modulated optical signals into a modulated output light beam.
In certain examples, a method includes obtaining, at a single sign-on (SSO) broker, an identity event associated with a user of a private cloud; requesting, by the SSO broker and in response to obtaining the identity event, a user information set corresponding to the user from a private cloud platform corresponding to the private cloud; providing the user information set to an identity access management (IAM) tool of the SSO broker; and updating a realm of the IAM tool to reflect the identity event based on the user information set.
Disclosed are a time synchronization processing method and apparatus. The method comprises: determining a time stability index of a node according to a time deviation value of the node within a preset period of time, where the time deviation value is a delay of a clock signal of the node relative to a reference clock signal within the preset period of time, and the time stability index comprises at least one of the following: a time compensation accumulated value within the preset period of time, a maximum time compensation value within the preset period of time, a time compensation average value within the preset period of time, and a time fluctuation value within the preset period of time; determining whether the time stability index of the node exceeds a preset range; and in a case that the time stability index of the node exceeds the preset range, sending a time synchronization exception alarm. In this manner, by means of the present invention, the problem that time stability of a network or device cannot be detected in the prior art is solved, so as to detect time stability of a network or device in real time according to a time stability index, and ensure time synchronization performance.
In some examples, a controller activates, as a response to an event, a power control signal to the processing resource through the interface, the activated power control signal to place the processing resource in a reduced power mode. The controller updates a value of a resource power capping parameter in an iterative power adjustment process. In an iteration of the iterative power adjustment process, the controller provides the updated value of the resource power capping parameter to the processing resource to set a power consumption cap of the processing resource.
The present disclosure proposes a solution of improving the TCP traffic latency for a station (STA) Multi-Link Device (MLD). The method can identify the TCP traffic of the STA MLD and obtain a traffic behavior of the STA MLD, and the traffic behavior of the STA MLD includes at least the size evaluation of the TCP traffic of the STA MLD, etc. Through collecting information of the number of available links and whether the link(s) of the STA MLD is proper for latency sensitivity traffic flow, the method can estimate the links of the STA MLD and obtain link health information. And the method can determine the TCP transmission mode for the STA MLD based on the traffic behavior and the link health information. Finally, the method transmits the traffic of the STA MLD based on the determined mode. In this way, the TCP traffic latency can be improved.
Example implementations relate to operations in a storage system. An example includes loading a container index into memory to match against new data units to be stored in a storage system. The example also includes, in response to loading the container index into the memory to match against the one or more new data units: reading metadata in the container index to identify a container entity group (CEG) object stored in the storage system; identifying a subset of unreferenced data units; in response to a determination that a size of the subset of unreferenced data units is greater than a threshold, storing a subset of referenced data units in a pending CEG object loaded in the memory; and after storing the subset of referenced data units in the pending CEG object, deleting the identified CEG object from the storage system.
Methods and systems for managing heterogeneous cloud resources are provided. The method includes normalizing heterogeneous data for provider-specific resources into a normalized data model that maintains relationships between components. A service request specifying an application is received. An orchestration sequence for the application is determined by analyzing the normalized data model to identify a subset of provider-specific resources for an infrastructure and day-2 environment of the application, and generating a process workflow. The orchestration sequence is executed by configuring the subset of provider-specific resources according to the process workflow. The method enables efficient orchestration of diverse resources across heterogeneous provider environments using a unified data model and workflow generation approach.
Methods and systems for managing heterogeneous cloud resources are provided. The method includes normalizing heterogeneous data for provider-specific resources into a normalized data model that maintains relationships between components. A service request is received, and an orchestration sequence is determined based on the request and the normalized data model. This involves analyzing the model to identify a subset of resources called for by the request and mapping dependencies to generate a process workflow. The orchestration sequence is then executed by configuring the subset of resources according to the workflow. The normalized data model enables efficient orchestration across heterogeneous environments while maintaining relationships and dependencies between resources.
In a network, a network device can receive, from a second network device, a notification comprising information associated with a join request for a multicast group received at the second network device. Here, the network device and the second network device can be rendezvous points (RPs) for the multicast group. The network device can determine, based on the notification, a first port of the network device corresponding to a second port of the second network device. The join request can be received at the second port from a device coupled to the first and second ports. The network device can then select the first port as an egress port for multicast traffic of the multicast group and, upon receiving a packet of the multicast traffic, send the packet to the device via the first port.
Systems and methods are provided to track and calculate the carbon generated by individual processing stages of a processing pipeline or an individual artifact. The system may track and store the lineage/dependencies of the data from individual applications that are involved in the processing pipeline. The applications may transmit their metadata to the system to record the metadata associated with the execution. The system can track the carbon footprint from start to finish of individual stages at the job/pipeline level or each of the processing steps of generating the individual artifact.
In some examples, a base filesystem layer stores a file. As part of provisioning a computer system, the computer system creates an upper filesystem layer that overlays the base filesystem layer, obscures the file to render at least a portion of the file inaccessible, and adds the obscured file to the upper filesystem layer. Responsive to an access request from a process targeting the file, the upper filesystem layer returns the obscured file to the process.
According to an implementation, a computer system manages application migrations between computing environments through reusable migration plans. A processor receives and stores expert-defined migration templates in a repository, retrieves compatible plans based on application characteristics, and executes migrations through standardized parameters. The system separates complex configuration from routine execution, enabling non-expert operators to implement migrations through simplified interfaces while maintaining operational consistency through automated validation and recovery capabilities.
G06F 9/455 - ÉmulationInterprétationSimulation de logiciel, p. ex. virtualisation ou émulation des moteurs d’exécution d’applications ou de systèmes d’exploitation
66.
NEUTRAL HOST NETWORKS FOR PRIVATE CELLULAR NETWORKS
Systems and methods provide for private cellular networks as Neutral Host Networks (NHN) by facilitating connections between a private cellular network and mobile network operator (MNO) core networks for authenticating user equipment (UEs) for access to the private cellular network using credentials for the MNO core networks. Examples include a connection system that receives an access request from a private cellular network that includes a UE identifier and a network identifier of an MNO, verifies that the MNO permits NHN services, and, based on the verification, establish a channel between the private cellular network and a core network corresponding to the network identifier. The connection system routes the access request message to the core network, which authenticates the UE. The UE can be granted access to the private cellular network based on the authentication from the core network.
In some examples, a system provides multi-chassis link aggregation by the first and second electronic devices that are part of a logical device supporting an MCLAG, where the first electronic device includes a first time clock, and the second electronic device includes a second time clock. The first and second time clocks perform, over a link between the first and second electronic devices of the logical device, a clock source selection process to select one of the first and second time clocks as a clock source and another one of the first and second time clocks as a clock sink as part of a time synchronization process in the system.
A system receives, by a network device in a network fabric, a packet comprising at least one of a traffic-class value for a traffic-class type and an endpoint-class value for an endpoint-class type. The system extracts the traffic-class value and the endpoint-class value from the packet and determines a hierarchical class structure, which indicates: priorities associated with types of classes, comprising a first priority associated with the traffic-class type and a second priority associated with the endpoint-class type; and bandwidth allocation ratios for values in a respective class type, comprising first bandwidth allocation ratios for values of the traffic-class type and second bandwidth allocation ratios for values of the endpoint-class type. The system determines a bandwidth allocation for the packet based on the extracted values and the hierarchical class structure and forwards the packet based on the determined bandwidth allocation for the packet.
Example implementations relate to deduplication operations in a storage system. An example includes receiving a set of data units to be stored in persistent storage of a deduplication storage system, and recording the arrival order of the set of data units in a manifest. The example also includes, in response to determining that an average unit size of the set of data units is smaller than a minimum size threshold, identifying a first data unit that is smaller than the minimum size threshold. The example also includes generating an aggregated data unit including the first data unit and a second data unit, and, in response to determining that the size of the aggregated data unit is larger than a target size threshold, replacing, in the manifest, the first data unit and the second data unit with the aggregated data unit.
Methods and systems for managing heterogeneous cloud resources are provided. Plugin interfaces are generated for service providers of provider-specific resources. Heterogeneous data for the provider-specific resources is normalized into a normalized data model by transforming resource definitions into defined schemas while maintaining relationships between components. A service request calling for a subset of the provider-specific resources is received. The subset of provider-specific resources is then orchestrated using the generated plugin interfaces and normalized data model. The plugin interfaces manage interactions with service providers to configure the requested resources according to the normalized model and service request.
Methods and systems for managing heterogeneous cloud resources are provided. A normalized data model is created by transforming provider-specific resource definitions into defined schemas while maintaining relationships between components. The normalized data model is stored in a database. Upon receiving a service request, a subset of provider-specific resources is orchestrated based on the normalized model. The database is then updated to reflect changes from the orchestration. This approach enables efficient management and orchestration of diverse cloud resources across multiple providers through a unified data model and orchestration process
G06F 9/50 - Allocation de ressources, p. ex. de l'unité centrale de traitement [UCT]
G06F 16/25 - Systèmes d’intégration ou d’interfaçage impliquant les systèmes de gestion de bases de données
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
72.
REUSABLE BARRIER FOR SYNCHRONIZATION AMONG MULTIPLE PROCESSES
One aspect provides a system and method for facilitating synchronization among processes. During operation, the system may execute, in parallel on one or more compute nodes, a plurality of processes. In response to a first process calling a barrier function, the system may pause execution of the first process, and in response to determining that the first process gains access to a variable shared by at least a subset of the plurality of processes, the system may update the shared variable. The system may release the shared variable to a second process in the subset to update the shared variable when the second process calls the barrier function and determine whether all processes in the subset have updated the shared variable. In response to the shared variable having been updated by all processes in the subset, the system may resume the execution of all processes in the subset.
Systems and methods are provided for performing anomaly detection. An example method includes, in a training phase, performing time series decomposition on training time series data to extract residuals of the training time series data, the residuals including a plurality of data points of the training time series data, using unsupervised anomaly detection models, identifying and labeling anomalous data points from among the plurality of data points contained in the residuals, based on outputs from the ensemble of unsupervised models including the labeled anomalous data points, obtaining a combined output indicating the labeled anomalous data points, and, using the combined output, training supervised anomaly detection models to detect anomalies in inference time series data In an inference phase, the method includes, using the trained ensemble of supervised anomaly detection models, on real-time, inference time series data.
H04L 41/16 - Dispositions pour la maintenance, l’administration ou la gestion des réseaux de commutation de données, p. ex. des réseaux de commutation de paquets en utilisant l'apprentissage automatique ou l'intelligence artificielle
74.
Avoiding unwanted duplication and/or flooding for layer 2 unknown unicast and/or layer 2 internet protocol (IP) multicast (BUM) traffic, for example, in devices used in VXLAN data centers and data center interconnects
The forwarding of BUM traffic in routers and/or switches (e.g., in environments such as interconnected L2 broadcast domains, such as DCs and DCIs with a VXLAN overlay) is improved so that BUM traffic is properly forward, but with reduced (e.g., eliminated) unnecessary duplication and/or flooding.
A network device may determine that a particular security association (SA) is to be established with a peer network device and that a post-quantum preshared key (PPK) cannot be obtained by the network device. The network device may determine that establishment of quantum-secure SAs is not mandatory for the network device. The network device may generate based on determining that the PPK cannot be obtained by the network device, and based on determining that establishment of quantum-secure SAs is not mandatory for the network device, a pseudo-PPK, and may generate, using the pseudo-PPK, pseudo-PPK authentication information associated with the network device. The network device may generate, using a non-quantum-secure key, non-quantum-secure authentication information associated with the network device. The network device may generate an authentication message that includes the pseudo-PPK authentication information and the non-quantum-secure authentication information, and may send the authentication message to the peer network device.
Systems and methods are provided for a model-centric approach that can be used to measure a computer's performance based on metrics obtained during and/or from training a machine learning (ML) model. Examples include building a training data set by generating first matrices and second matrices and deriving third matrices from the first and second matrices. Examples also include training, at a plurality of computer systems, a plurality of machine learning (ML) models by applying the first and third matrices to a plurality of ML algorithms and obtaining performance metrics based on training. The performance metrics can then be set as benchmarks for the plurality of computing systems to facilitate assessing a relative performance amongst the plurality of computing systems.
In example implementations, a computer system includes first memory for file storage and second memory storing a first mapping table and a second mapping table. The first mapping table associates user addresses with fingerprints and the second mapping table associates the fingerprints with storage locations of the first memory. Instructions cause one or more processors to receive an encrypted data file associated with a received user address and a fingerprint associated with the received encrypted data file. Based on the received fingerprint, it is determined whether the received encrypted data file is a duplicate of a previously stored data file. If the received encrypted data file is not a duplicate, the received encrypted data file is stored in the first memory and the first and second mapping tables are updated. If the received file is a duplicate, the first mapping table is updated.
A region-aware GPU power/energy regulation method comprises periodically identifying a phase of execution of an application which is currently being executed by a GPU and measuring the utilization of the GPU (e.g., memory utilization) during execution of the identified phase. The utilization may be measured during a sampling period at both high and low GPU frequencies. A frequency sensitivity parameter is then determined for the identified phase based on the measured utilization of the GPU. A selected frequency for the identified phase is then determined based on the frequency sensitivity parameter. The GPU can then be instructed to set a frequency of the GPU to the selected frequency during execution of the remainder of the identified phase.
In some examples, a system receives a representation of a honeypot pattern and information of a honeypot object containing the honeypot pattern injected into primary data. The system checks backup data created by a backup management system by identifying an instance of the honeypot object in the backup data, and determining whether data of the instance of the honeypot object deviates from the honeypot pattern. Based on determining that the data of the instance of the honeypot object deviates from the honeypot pattern, the system triggers a remediation action relating to the backup data.
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
In some examples, a distributed system assigns key-value pairs to respective compute nodes of a plurality of compute nodes based on relationships of key identifiers of keys in the key-value pairs and node identifiers of the respective compute nodes on an identifier ring. A compute node determines whether a first gap on the identifier ring between node identifiers of first successive compute nodes is larger than a second gap on the identifier ring between node identifiers of second successive compute nodes. Based on determining that the first gap is larger than the second gap, the compute node initiates a shift operation that changes a node identifier of the compute node of the first successive compute nodes to reduce a size of the first gap on the identifier ring.
H04L 67/1008 - Sélection du serveur pour la répartition de charge basée sur les paramètres des serveurs, p. ex. la mémoire disponible ou la charge de travail
H04L 67/1025 - Adaptation dynamique des critères sur lesquels repose la sélection du serveur
82.
ORDERING OF BAY IDENTIFIERS IN AN ELECTRONIC MODULE
In some examples, an electronic module includes a moveable contact engageable with a housing of a system based on a mounting orientation of the electronic module in the system. The electronic module includes a plurality of bays to receive devices, and a memory including a first memory location storing identifiers of the plurality of bays in a first order, and a second memory location storing the identifiers of the plurality of bays in a second order different from the first order. The electronic module includes a switch assembly to selectively select the first memory location or the second memory location based on a position of the moveable contact.
Systems and methods are provided for automatically generating an embedding that is linked to a newly created data object by its primary key. For example, in response to entering a data object into an object data structure of the object storage system, the system may automatically generate an embedding comprising a primary key of the data object that links the embedding with the data object. In response to receiving an update of the data object, the system may automatically identify the primary key of the data object and synchronize, using a notification service of the object storage system, the embedding with the update of the data object.
A heterogenous probabilistic computer architecture comprises a probabilistic processing unit (PPU), a central processing unit (CPU), a graphics processing unit (GPU), and a bus communicably connecting the PPU, CPU, and GPU. A heterogenous probabilistic computer using this architecture may form a sampling and optimization problem solver configured to process a sampling and optimization workload, such as an energy based model (EBM). In processing the sampling and optimization workload, the PPU may be used to generate samples, while the GPU may be used to compute gradients, weights, biases and/or other values related to the samples. The PPU and the GPU may communicate directly with one another using peer-to-peer communications via the bus. A quantum processing unit (QPU) may also be used, in some examples, to accelerate sampling.
G06N 10/40 - Réalisations ou architectures physiques de processeurs ou de composants quantiques pour la manipulation de qubits, p. ex. couplage ou commande de qubit
G06N 10/60 - Algorithmes quantiques, p. ex. fondés sur l'optimisation quantique ou les transformées quantiques de Fourier ou de Hadamard
Provided herein are techniques for workflow task assignment in a hybrid implementation environment having a plurality of different implementation environments with differing implementation benefits. A workflow of tasks is analyzed to identify a first subset of tasks to be implemented in a first one of the implementation environments and a second subset of tasks to be implemented in a second implementation environment. The first subset and second subset are determined based upon relevant characteristics of the implementation, such as characteristics of the workflow, tasks, and/or implementation environment(s).
Server racks may include an enclosure with an airflow inlet in a first lateral sidewall and an airflow outlet in a second, opposite lateral sidewall. A support structure may be rotatably mounted to the enclosure and rotatable between an open position and a closed position. Various other apparatuses, systems, and methods are also disclosed.
In an example implementation, a network device includes first switch circuitry associated with a first set of physical ports and second switch circuitry associated with a second set of physical ports. The first and second sets of physical ports may collectively form a combined set of physical ports. A communication link may couple the first switch circuitry and the second switch circuitry. A mapping of the first set of physical ports and the second set of physical ports to single set of virtual ports may be stored. One or more processors are configured to receive a transmission on a first physical port of the combined set of physical ports, access the mapping to determine a first virtual port of the single set of virtual ports that corresponds to the first physical port, and generate an indication of the first virtual port as a communication port for the transmission.
H04L 41/08 - Gestion de la configuration des réseaux ou des éléments de réseau
H04L 41/0816 - Réglages de configuration caractérisés par les conditions déclenchant un changement de paramètres la condition étant une adaptation, p. ex. en réponse aux événements dans le réseau
H04L 41/0893 - Affectation de groupes logiques aux éléments de réseau
H04L 45/586 - Association de routeurs de routeurs virtuels
88.
MIGRATION FOR NETWORK-BASED VIRTUAL MACHINE REPLICATION
A method and system for replicating data change operations in a virtualized environment is provided. A data change filter and a data change driver in a hypervisor intercept data change operations from a virtual machine. A replication processing service receives a first stream of data change operations from the data change driver and a second stream from the data change filter. The service identifies a matching data change operation in both streams, replicates operations from the first stream up to a transition point, and then replicates operations from the second stream starting from the transition point. The transition point is pre-defined with respect to the matching data change operation. This approach enables seamless migration between driver-based and filter-based replication methods while maintaining data consistency and continuity.
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
A method and system for replicating data change operations in a virtualized environment is provided. A data change filter in a hypervisor of a virtualization host intercepts data change operations from a virtual machine. A network connection is established between the data change filter and a replication processing service executing on a separate replication host. The replication processing service receives the data change operations from the data change filter over the network connection and replicates the data change operations to a backup site.
G06F 9/455 - ÉmulationInterprétationSimulation de logiciel, p. ex. virtualisation ou émulation des moteurs d’exécution d’applications ou de systèmes d’exploitation
90.
AUTHENTICATION FOR NETWORK-BASED VIRTUAL MACHINE REPLICATION
A method and system for configuring a data change filter in a virtualized environment are provided. A data change filter is installed in a hypervisor of a virtualization host, where the hypervisor executes a virtual machine. The data change filter intercepts data change operations from the virtual machine. The hypervisor includes a certificate management service that stores a private certificate for the data change filter and a public certificate for a replication processing service. The data change filter retrieves the certificates from the certificate management service, establishes an authenticated network connection with the replication processing service using the certificates, and sends the intercepted data change operations to the replication processing service over the authenticated connection. The system enables secure replication of data changes in virtualized environments.
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
G06F 9/455 - ÉmulationInterprétationSimulation de logiciel, p. ex. virtualisation ou émulation des moteurs d’exécution d’applications ou de systèmes d’exploitation
91.
USING NOISE IN MEMRISTORS FOR DIFFERENTIAL PRIVACY
In certain examples, a method may include receiving a privacy parameter and selecting an electrical property range for cells in a crossbar array based on the privacy parameter. The cells in the crossbar array may then be programmed based on the selected electrical property range, which may provide a certain level of differential privacy.
Methods and systems relate to packet forwarding that includes a network device receiving a message from a second network device and that has a destination address of a third network device. The circuitry of the network device determines that a connection between the second network device and the third network device has been previously opened by the third network device. In response to the determination that the connection is open and in response to receiving the message, the circuitry transmits the message towards the third network device.
A system determines a set of workflow languages which capture tasks to be executed in a corresponding workflow. The system defines a set of classes of expressivity, wherein a class of expressivity represents a workflow language. The system identifies, in the set of workflow languages, an input language and a target output language. The system determines whether the target output language is a match for the input language by comparing a respective class of expressivity for the input language and the respective class of expressivity for the target output language. The system returns information associated with whether the target output language is a match for the input language.
In a network, a network device can determine a role for a client device coupled to a first port of the network device. The network device can then determine a set of telemetry parameters associated with the role. The network device can also determine a flow identifier of a data flow received from the client device via the first port. Subsequently, the network device can identify a respective packet in the data flow associated with the flow identifier. The network device can then record, based on a telemetry process of the network device, the set of telemetry parameters associated with the packet.
In certain examples, a method includes detecting, by a network device, that a computing device at an edge site requires provisioning; obtaining, by the network device, identifying information corresponding to the computing device from the computing device; providing, by the network device, the identifying information to a cloud service device; receiving, from the cloud service device and at a proxy provisioning container executing on the network device, provisioning information corresponding to the computing device; obtaining, by the proxy provisioning container, provisioning items based on the provisioning information; and orchestrating, by the proxy provisioning container, provisioning of the computing device using the provisioning items.
A technique includes monitoring health metric values associated with a collection of monitored resources associated with a microservice. The technique includes determining based on the health metric values, whether each resource of the collection of monitored resources is healthy or unhealthy. The determination of whether each resource is healthy or unhealthy includes determining that a given resource of the collection of resources is healthy. The technique includes for each resource of the collection of resources, monitoring an associated security status of the resource; and determining availability statuses for the collection of resources. Determining the availability statuses includes classifying each resource that is unhealthy as being unavailable and classifying the given resource as being unavailable responsive to the security status associated with the given resource. The technique includes determining a resource availability of the microservice based on the availability statuses.
Systems and methods are provided for optically implemented Kolmogorov-Arnold Networks (KAN). Examples provide a photonic Kolmogorov-Arnold Network that includes a plurality of neurons and a plurality of synaptic edges. Each synaptic edge comprises a waveguide that optically couples a neuron of the plurality to another neuron of the plurality of neurons, and a nonlinear optical modulator formed on the waveguide, wherein the nonlinear optical modulator is configured to be tuned to a desired nonlinear activation function.
G06N 3/067 - Réalisation physique, c.-à-d. mise en œuvre matérielle de réseaux neuronaux, de neurones ou de parties de neurone utilisant des moyens optiques
Systems and methods are provided for creation of project meshes by defining agentic tool instances that can be activated to access agentic tools. Examples include providing an interface for defining a mesh of agentic tools and receiving, via the interface, queries descriptive of a project. The queries can be provided to a reasoning engine that generates first parameters defining tasks and second parameters defining agentic tool instances. The agentic tool instances can be created in accordance with the first parameters and the second parameters and activated for accessing agentic tools located in a data store. The agentic tools can correspond to a parameter of the first parameters associated an agentic tool instance. Examples can configure the mesh comprising the agentic tool instances, and the mesh can be executed to cause the agentic tool instances to access and run one or more of the agentic tools.
A device may receive, from an organization user, a request for support associated with a cloud computing environment utilized by the organization user, and may provide the request for support to a support user. The device may receive, from the support user, credentials of the support user and a login request to access the cloud computing environment, and may determine whether the credentials of the support user satisfy a domain check, a virtual private network (VPN) check, a role check, and a secure group check. The device may selectively deny the login request based on the credentials failing to satisfy one or more of the domain check, the VPN check, the role check, or the secure group check, or may approve the login request based on the credentials satisfying the domain check, the VPN check, the role check, and the secure group check.
Example implementations relate to deduplication operations in a storage system. An example includes receiving a data stream that includes multiple locality portions. The example also includes, upon determining that a first container index for a first portion has insufficient capacity to store metadata of a first set of new data units in the first portion, instantiating a first overflow container index associated with the first portion, and storing the metadata of the first set of new data units in the first overflow container index. The example also includes, upon determining that a second container index associated with a second portion has insufficient capacity to store metadata of a second set of new data units in the second portion, instantiating a second overflow container index associated with the second portion, and storing the metadata of the second set of new data units in the second overflow container index.
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 16/215 - Amélioration de la qualité des donnéesNettoyage des données, p. ex. déduplication, suppression des entrées non valides ou correction des erreurs typographiques
brevets
