G06F11/30

Electronic element, system comprising such an electronic element and method for monitoring and cutting off a processor on occurrence of a failure event
11579995 · 2023-02-14 · ·

An electronic element includes: a module for storing reference data; a module for receiving data from a processor; a module for verifying the received data by comparison by way of reference data; and a module for transmitting an instruction to cut off supply of the processor, the supply cutoff instruction being transmitted after occurrence of a failure event, the failure event being an absence of reception of data or a failure in verifying the data. A system including such an electronic element and a method for monitoring a processor by the electronic element are also described.

Dynamic graphical processing unit register allocation

Systems, apparatuses, and methods for dynamic graphics processing unit (GPU) register allocation are disclosed. A GPU includes at least a plurality of compute units (CUs), a control unit, and a plurality of registers for each CU. If a new wavefront requests more registers than are currently available on the CU, the control unit spills registers associated with stack frames at the bottom of a stack since they will not likely be used in the near future. The control unit has complete flexibility determining how many registers to spill based on dynamic demands and can prefetch the upcoming necessary fills without software involvement. Effectively, the control unit manages the physical register file as a cache. This allows younger workgroups to be dynamically descheduled so that older workgroups can allocate additional registers when needed to ensure improved fairness and better forward progress guarantees.

Methods and systems parallel raid rebuild in a distributed storage system

A method for rebuilding data, comprising: obtaining, from a metadata node, a source file data layout for a source file and a target file data layout for a target file, wherein the source file is associated with a degraded mapped RAID group and the target file is associated with a new mapped RAID group; generating, by the client application node, a plurality of input/output (I/O) requests to read a portion of the data associated with the source file using the source file data layout; obtaining, in response to the plurality of I/O requests, the portion of the data associated with the source file; rebuilding a second portion of the data associated with source file using the portion of the data; and initiating, storage of at least the second portion of the data associated with the source file in the storage pool using the target file data layout.

Reducing save restore latency for power control based on write signals

A method of save-restore operations includes monitoring, by a power controller of a parallel processor (such as a graphics processing unit), of a register bus for one or more register write signals. The power controller determines that a register write signal is addressed to a state register that is designated to be saved prior to changing a power state of the parallel processor from a first state to a second state having a lower level of energy usage. The power controller instructs a copy of data corresponding to the state register to be written to a local memory module of the parallel processor. Subsequently, the parallel processor receives a power state change signal and writes state register data saved at the local memory module to an off-chip memory prior to changing the power state of the parallel processor.

Method for managing multiple operating systems in a terminal

The disclosure provides a method for managing multiple operating systems in a terminal. The terminal includes multiple operating systems and a management system. The management system is configured to manage the multiple operating systems. The management system includes a cross-system application database. The method includes: when a first operating system in the multiple operating systems runs in a foreground, and a second operating system in the multiple operating systems runs in a background, if the second operating system receives a first message of a first application in the second operating system, sending, by the second operating system, a notification message to the management system; storing, by the management system, the notification message into the cross-system application database; and listening, by the first operating system, on the cross-system application database, and outputting a prompt of the first message when listening and obtaining the notification message.

Merging scaled-down container clusters using vitality metrics
11579935 · 2023-02-14 · ·

A system for container migration includes containers running instances of an application running on a cluster, an orchestrator with a controller, a memory, and a processor in communication with the memory. The processor executes to monitor a vitality metric of the application. The vitality metric indicates that the application is in either a live state or a dead state. Additionally, horizontal scaling for the application is disabled and the application is scaled-down until the vitality metric indicates that the application is in the dead state. Responsive to the vitality metric indicating that the application is in the dead state, the application is scaled-up until the vitality metric indicates that the application is in the live state. Also, responsive to the vitality metric indication transitioning from the dead state to the live state, the application is migrated to a different cluster while the horizontal scaling of the application is disabled.

Systems and methods for dynamic aggregation of data and minimization of data loss
11579999 · 2023-02-14 · ·

A computer-implemented system for dynamic aggregation of data and minimization of data loss is disclosed. The system may be configured to perform instructions for: aggregating information from a plurality of networked systems by collecting a set of data from the networked systems, the set of data comprising data associated with a predetermined period of time and comprising one or more central variables that are included in data associated with more than one networked systems of the plurality of networked systems and one or more associated variables that describe one or more aspects of the central variables; retrieving one or more data transformation rules based on a relational map among the central variables and the associated variables; and aggregating the first set of data into one or more master data structures corresponding to the central variables based on the data transformation rules.

Optimizing host CPU usage based on virtual machine guest OS power and performance management

Techniques for optimizing CPU usage in a host system based on VM guest OS power and performance management are provided. In one embodiment, a hypervisor of the host system can capture information from a VM guest OS that pertains to a target power or performance state set by the guest OS for a vCPU of the VM. The hypervisor can then perform, based on the captured information, one or more actions that align usage of host CPU resources by the vCPU with the target power or performance state.

Performance monitoring in a distributed storage system
11582130 · 2023-02-14 · ·

Methods and systems for monitoring performance in a distributed storage system described. One example method includes identifying requests sent by clients to the distributed storage system, each request including request parameter values for request parameters; generating probe requests based on the identified requests, the probe requests including probe request parameter values for probe request parameter values, representing a statistical sample of the request parameters included in the identified requests; sending the generated probe requests to the distributed storage system over a network, wherein the distributed storage system is configured to perform preparations for servicing each probe request in response to receiving the probe request; receiving responses to the probe requests from the distributed storage system, and outputting at least one performance metric value measuring a current performance state of the distributed storage system based on the received responses.

Memory device with configurable performance and defectivity management

A memory device comprises a memory control unit including a processor configured to control operation of the memory array according to a first memory management protocol for memory access operations, the first memory management protocol including boundary conditions for multiple operating conditions comprising program/erase (P/E) cycles, error management operations, drive writes per day (DWPD), and power consumption; monitor operating conditions of the memory array for the P/E cycles, error management operations, DWPD, and power consumption; determine when a boundary condition for one of the multiple operating conditions is met; and in response to determining that a first boundary condition for a first monitored operating condition is met, change one or more operating conditions of the first memory management protocol to establish a second memory management protocol for the memory access operations, the second memory management protocol including a change boundary condition of a second monitored operating condition.