Abstract: A processing system includes a driver and an accelerated processing unit including a processor. The processor is configured to initiate a status check of wavefronts being executed by the accelerated processing unit responsive to receiving a status inquiry from the driver. Responsive to the status check indicating a hang, the processor is configured to employ a machine learning algorithm to selectively extract data from one or more registers of the accelerated processing unit. For example, in some cases, the one or more registers are local to one or more compute units of the accelerated processing unit. The processor is further configured to export the data from the accelerated processing unit prior to the accelerated processing unit being reset.

Abstract: Providing adaptive and dynamic redundancy for functional safety in processor devices is disclosed herein. In some aspects, a processor device comprises a central configurable redundancy logic block (CRLB) controller circuit, a redundancy map matrix switch circuit, a plurality of safety-critical block circuits, and a plurality of CRLB clusters. The central CRLB controller circuit receives redundancy mapping data for each safety-critical block circuit, and receives an indication of one or more active block circuits among the safety-critical block circuits. The central CRLB controller circuit transmits the redundancy mapping data corresponding to the active block circuits to the redundancy map matrix switch circuit, which configures one or more CRLB clusters to duplicate functionality of the respective active block circuits based on the redundancy mapping data. The redundancy map matrix switch circuit then provides logic redundancy for the one or more active block circuits using the one or more CRLB clusters.

Abstract: Methods, systems, and computer-readable storage media for receiving, by a management tool of a software vendor, a payload including system data that is representative of a software system experiencing an issue, generating, by the management tool, a service using the system data provided in the payload, executing the service to trigger an automated procedure that collects one or more log files from one or more source systems, uploads the one or more log files to a repository, and provides a uniform resource locator (URL) that provides a link to the one or more log files within the repository, and transmitting a response to a ticketing system, the response including the URL.

Abstract: Methods and apparatus for maintaining the cooling systems of distributed compute systems are disclosed. An example apparatus disclosed herein includes memory, machine-readable instructions, and processor circuitry to execute the machine-readable instructions to determine a health of a server, determine a threshold based on a workload service level agreement associated with the server, and in response to determining the health does not a satisfy the threshold, throttle a workload on the server.

Abstract: The present disclosure describes a method and apparatus for an error remediation recommendation process based on multiple data sources. Aspects of the disclosure may provide for processing of a knowledgebase to identify error remediation recommendations, processing of a chat log to identify error remediation recommendations, identification of errors from log data, determination of a corresponding error remediation recommendation from the knowledgebase and/or the chat log, and causing display of output indicating a corresponding error remediation recommendation. Further aspects described may also provide for training of an artificial neural network to determine a corresponding error remediation recommendation. Further aspects described may also provide for a browser extension which scrapes the Document Object Model (DOM) of a web page to identify errors.

Abstract: The present description concerns a bus error management method, wherein one or a plurality of first characteristics of a first write transaction intended for a functional unit and transiting through a bridge, are stored, and wherein in the presence of a bus error sent by the functional unit: one or a plurality of second characteristics linked to said error are stored; the bridge generates a first interrupt that it transmits with said first and second characteristics to a management unit; and the management unit generates at least one second interrupt intended for a processing unit as a function of the first and/or second characteristics.

Abstract: By detecting panic situations and providing a host device detailed panic data and mitigation options before the panic situation condition occurs, failure of data storage devices may be reduced. Several mitigation options may be presented to the host device upon detection of a future panic situation, such as adjusting read performance; increasing device power; performing evacuation and management operations; and/or redirecting host commands to another data storage device for command completion. When the other data storage device is in the same PCIe tree and reachable, the command may be directed with PRPs/SGLs that point back to the same host device. In some embodiments, the data storage devices may have a submission queue between them. In some embodiments, the data storage device comprises a panic early detection module and a panic control module.

Abstract: Techniques for monitoring the health of services of a system are disclosed. A system determines that a detected alarm is associated with a service feature, and the service feature is associated with a service of a cloud environment. The system computes a health metric for the service based at least on the detected alarm that is associated with the service feature. Additionally, the system generates a visual representation that includes the health metric for display on a service health interface.

Abstract: A processing device in a host system detecting a failure of a memory sub-system coupled to the host system and collects debug information from the memory sub-system via a memory sub-system interface. Responsive to determining that the debug information is not sufficient to correct the failure of the memory sub-system, the processing device performing an authentication process to authenticate the system to the memory sub-system, collects enhanced debug information from the memory sub-system via the memory sub-system interface, and performs one or more corrective actions based on the enhanced debug information to correct the failure of the memory sub-system.

Abstract: In a storage system including a plurality of storage nodes and a plurality of storage devices that provide storage areas to the storage nodes, a cluster control unit that has detected occurrence of a failure in another storage node requests an external control device to execute a detach process of detaching a storage device allocated to a failure storage node from the failure storage node and an attach process of allocating the storage device to an alternative storage node. Then, the alternative cluster control unit restores, based on a log stored in a storage device allocated to a storage node in which a most recent failure has occurred among the storage nodes belonging to the same redundancy group as a failure storage control unit of the failure storage node, storage content of a memory of the alternative storage node.

Abstract: A method may include: receiving a plurality of log-text examples from network devices and labels indicating if the log-text example is anomalous or not anomalous; identifying, from the log-text examples and labels, rules for weak annotation, and validation labels for a validation dataset and test labels for a test dataset; organizing log-text examples in the validation dataset into a plurality of time windows; concatenating log-text examples in each time window; providing the concatenated log-text examples and the rules for weak annotation to a weak annotation framework, to a LLM training dataset; training a LLM with the LLM training dataset; collecting runtime log-texts from the network devices for a period of time; concatenating the runtime log-texts for the period of time; and prompting the LLM for analysis with the concatenated log-texts, wherein the LLM outputs a response of anomaly or no anomaly and a confidence in the response.

Abstract: Upon failure of a host and, in response to a lack of hots having available processing units, a host is selected and one or more processing units of the selected host are allocated as shared CPUs for use by one or more components of the failed host. The selected host may be selected according to requirements, such as affinity, anti-affinity, and latency. The shared CPUs may have been previously allocated as a dedicated CPU. The shared CPUs may be bound to the one or more components. The one or more components may include a container.

Abstract: Architectures and techniques are described that can leverage a container storage interface (CSI) driver to initiate volume health tests for a storage system in accordance with certain embodiments of this disclosure. The CSI driver can conform to a CSI specification or standard and can be used to expose the storage system to a container orchestration system. The volume health tests can be tailored to reduce or mitigate the impact of testing based on a status or state of the volumes. For instance, enforcement can be provided to ensure that only shorter, low-impact testing is performed on volumes having a published state, whereas longer, more intensive testing can be performed on volumes that are not in the published state.

Abstract: In some examples, a system detects disabling of a driver of a storage control feature included in a main processor of the system, where the storage control feature to manage access of a storage device. In response to detecting the disabling of the driver of the storage control feature included in the main processor, the system initiates a remediation action to prevent a fault in the system.

Abstract: A method for reporting a hot-swap error, a processor architecture, a device, and a storage medium are provided by an embodiment of the present application. The method is applied to an advanced reduced instruction-set computer (RISC) machine (ARM) platform, the ARM platform includes a baseboard management controller (BMC), an operating system (OS), and a basic input output system (BIOS). Data transmission between the BMC and the OS is performed based on a data interface service configured via the BIOS. During operation of the ARM platform, in response to a system control processor (SCP) in the BIOS detecting a PCIe CE type reporting error information corresponding to any interface, the SCP transmits the PCIe CE type reporting error information to the BMC and the OS.

Abstract: Methods, systems, and devices for reset techniques for protocol layers of a memory system are described. A communications link may be established between a host system and the memory system. In some examples, the communications link may be based on one or more first parameters associated with a first protocol layer and one or more second parameters associated with a second protocol layer. The system may support communication (e.g., from the host system to the memory system) of an indication to reset the communications link, and the host system, the memory system, or both may reset the one or more first parameters based on communicating the indication to reset the communications link. The host system and memory system may attempt to reestablish the communications link based on resetting the one or more first parameters and maintaining the one or more second parameters.

Abstract: A system can maintain a data storage system, wherein the data storage system comprises persistent memory and chunk storage, wherein the persistent memory is more performant than the chunk storage, wherein the chunk storage stores data in chunks of a fixed-size, and wherein respective chunks of the chunks correspond to respective parity values. The system can, based on receiving a request associated with a storage account to write data to the data storage system, write the data to the persistent memory concurrently with writing the data to the chunk storage. The system can, after completing writing the data to the persistent memory and before completing writing the data to the chunk storage, send an indication, to the storage client, that the data has been successfully stored in the data storage system.

Abstract: Systems and methods for in-system, self-recovery (ISSR) in a system-on-a-chip (SoC) are disclosed for detecting and recovering from a bit corruption event that has caused an SoC subsystem to crash. If a bit flip event is detected through observation of a subsystem crash, ISSR steps are taken in an attempt to correct the issue. If the ISSR steps are successful, then the system is not halted and mission mode operations continue. If the ISSR steps are unsuccessful, then the system is halted and an return material authorization (RMA) can then be issued for return of the failed part to the OEM. Crash events and self-recovery attempts preferably are logged by the ISSR system and provided to the OEM.

Abstract: The present disclosure provides a method for processing storage medium failure, an apparatus, and a solid state disk thereof. The method includes: acquiring a failed storage medium; detecting a state of the failed storage medium in real time, where the state at least includes an idle state, an open state, and a closed state; and processing the failed storage medium in real time according to the state of the failed storage medium.

Abstract: Embodiments of the present disclosure disclose a method and a system for managing geo-redundant cloud servers. The method includes: receiving, by a set of first geo-redundant operators associated with a set of first databases hosted on a set of first servers of a first data system, state information of corresponding first server of the set of first servers, health information of a set of second servers of a second data system, from a corresponding first manager of a set of first managers associated with the set of first servers. The method includes synchronizing, by the set of first geo-redundant operators, stateful data from each first database of the set of first databases to a corresponding second database of a set of second databases associated with the set of second servers based on the state information, the health information and one or more synchronization parameters.