Techniques for managing resources of a virtualized system involve acquiring historical distribution data about a virtualized system, the historical distribution data indicating a historical distribution of resources occupied by workloads on a plurality of host machines of the virtualized system over a predetermined historical time period. The techniques further involve generating predicted distribution data based on the historical distribution data, the predicted distribution data indicating an estimated distribution of resources occupied by the workloads on the plurality of host machines over a predetermined future time period. The techniques further involve performing workload migration at least once based on the predicted distribution data, the workload migration including migrating a workload of a first host machine whose first estimated quantity of occupied resources exceeds a high threshold to a second host machine whose second estimated quantity of occupied resources is below a low threshold.

A cloud-based migration system exposes a source-independent application programming interface for receiving data to be migrated. The data is uploaded and stored as a single entity in a cloud-based storage system. A migration system then accesses the migration package and begins migrating the data to its destination, from the cloud-based storage system.

A method for distributed neural network processing, the method may include detecting, by a local neural network that belongs to a local device, and based on sensed information, an occurrence of a triggering event for executing or completing a classification or detection process; sending to a remote device, a request for executing or completing the classification or detection process by a remote device that comprises a remote neural network; wherein the remote neural network has more computational resources than the local neural network; determining by the remote device whether to accept the request; and executing or completing, by the remote device, the classification or detection process when determining to accept the request; wherein the executing or completing involves utilizing the remote neural network.

A task in an application node is managed. For instance, based on a type of a predetermined task that is to be executed on a data object in the application node, an address range of a group of objects on which the predetermined task is to be executed is determined in the data object. The predetermined task is executed on the group of objects in an order of addresses of the group of objects. A progress indicator is created for indicating an address of an object that is currently being processed in the group of objects. The predetermined task is managed based on the progress indicator. Thus, an address of an object that is currently being processed may be indicated based on the progress indicator, so that the predetermined task may be managed more easily and effectively based on the progress indicator in subsequent operations.

Systems and methods are provided for offloading a task from a central processor in a radio access network (RAN) server to one or more heterogeneous accelerators. For example, a task associated with one or more operational partitions (or a service application) associated with processing data traffic in the RAN is dynamically allocated for offloading from the central processor based on workload status information. One or more accelerators are dynamically allocated for executing the task, where the accelerators may be heterogeneous and may not comprise pre-programming for executing the task. The disclosed technology further enables generating specific application programs for execution on the respective heterogeneous accelerators based on a single set of program instructions. The methods automatically generate the specific application programs by identifying common functional blocks for processing data traffic and mapping the functional blocks to the single set of program instructions to generate code native to the respective accelerators.

A processor may identify an action of a first device of the IoT devices. The processor may initiate a transfer from the first device to one or more other devices of the IoT devices. The processor may pause each thread being executed by the first device and the one or more other devices. The processor may transfer a unit from the first device to the one or more other devices.

Techniques are provided for scheduling multiple jobs on one or more cloud computing instances, which provide the ability to select a job for execution from among a plurality of jobs, and to further select a designated instance from among a plurality of cloud computing instances for executing the selected job. The job and the designated instance are each selected based on a probability distribution that a cost of executing the job on the designated instance does not exceed the budget. The probability distribution is based on several factors including a cost of prior executions of other jobs on the designated instance and a utility function that represents a value associated with a progress of each job. By scheduling select jobs on discounted cloud computing instances, the aggregate utility of the jobs can be maximized or otherwise improved for a given budget.

This application provides an application program management method and apparatus. The method is performed in a database cluster system including at least two database nodes, at least one database object is stored in each database node, and the method includes: running an application program on a first database node in a first time period; determining a target database node based on at least one historical database object accessed by the application program in the first time period, where the target database node stores the historical database object; and running the application program on the target database node in a second time period. According to this application, a database node on which an application program runs can be dynamically adjusted, to avoid overload of the database node.

An instruction offload manager receives, by a processing device, a first request to execute a program, identifies one or more instructions of the program to be offloaded to a second processing device, where the second processing device includes a same instruction set architecture as the processing device, and provides the one or more instructions to a memory module comprising the second processing device. Responsive to detecting an indication to execute the one or more instructions, the instruction offload manager provides an indication to the second processing device to cause the second processing device to execute the one or more instructions, the one or more instructions to update a portion of a memory space associated with the memory module.

Embodiments described herein relate to a method for new endpoint addition. The method may include receiving, during execution of a migration workflow, a request to add a new endpoint to the migration workflow. The execution of the migration workflow includes performing a first migration job associated with a first consistency group and assigned a first priority; making a first determination that the first priority is higher than a priority threshold; based on the first determination, completing the first migration job; performing, after completing the first migration job, a new endpoint addition action set. The method may also include adding, based on the new endpoint migration job priority, the new endpoint migration job to a queue of remaining migration jobs of the migration workflow.