A process control method and apparatus are provided to implement intelligent control on each process in a terminal device and prevent a process important to a user from being killed while memory is released. The method includes: obtaining information, a package name, and a process name that are of a first process of a first application running on a terminal device; based on the information about the first process, determining to kill the first process; marking the package name and the process name that are of the first process; and based on marking results of a package name and a process name that are of a second process, controlling the second process.

A system for allocation of resources and processing jobs within a distributed system includes a processor and a memory coupled to the processor. The memory includes at least one process and at least one resource allocator. The process is adapted for processing jobs within a distributed system which receives jobs to be processed. The resource allocator is communicably coupled with at least one process, and is adapted to generate one or more sub-processes within a limit of one or more resources allocated to the process for processing jobs.

Remote kernel execution in a heterogeneous computing system can include executing, using a device processor of a device communicatively linked to a host processor, a device runtime and receiving from the host processor within a hardware submission queue of the device, a command. The command requests execution of a software kernel and specifies a descriptor stored in a region of a memory of the device shared with the host processor. In response to receiving the command, the device runtime, as executed by the device processor, invokes a runner program associated with the software kernel. The runner program can map a physical address of the descriptor to a virtual memory address corresponding to the descriptor that is usable by the software kernel. The runner program can execute the software kernel. The software kernel can access data specified by the descriptor using the virtual memory address as provided by the runner program.

Systems and methods related to integrated memory pooling and direct swap caching are described. A system includes a compute node comprising a local memory and a pooled memory. The system further includes a host operating system (OS) having initial access to: (1) a first swappable range of memory addresses associated with the local memory and a non-swappable range of memory addresses associated with the local memory, and (2) a second swappable range of memory addresses associated with the pooled memory. The system further includes a data-mover offload engine configured to perform a cleanup operation, including: (1) restore a state of any memory content swapped-out from a memory location within the first swappable range of memory addresses to the pooled memory, and (2) move from the local memory any memory content swapped-in from a memory location within the second swappable range of memory addresses back out to the pooled memory.

In an example, a computer-implemented method for deploying a workload in a virtualized computing environment include retrieving a digest file corresponding to the workload. The digest file may include a plurality of hash values from a storage device and each hash value corresponds to a data block of a plurality of data blocks associated with a virtual disk stored in the storage device. Further, the method includes determining whether the plurality of hash values in the digest file match with data in a CBRC of a destination host computing system and obtaining data blocks corresponding to hash values that are not present in the CBRC from the storage device to store in the destination host computing system. Furthermore, the method includes deploying the workload on the destination host computing system upon obtaining the data blocks corresponding to hash values that are not present in the CBRC.

An illustrative computing system may include a master node that includes a prioritization module. The master node may be in communication with each of a plurality of containerized application nodes. The prioritization module may be configured to determine node prioritization information indicating a relative prioritization of the containerized application nodes for instantiating a designated containerized application, wherein the determining node prioritization information includes assigning a priority score to a candidate containerized application node, wherein higher priority is given based on a number of volumes stored on the candidate containerized application node that is designated for access by the designated containerized application.

Described are self-learning systems and methods for adaptive management of memory resources within a memory hierarchy. Memory allocations associated with different active functions are organized into blocks for placement in alternative levels in a memory hierarchy optimized for different metrics of e.g. cost and performance. A host processor monitors a performance metric of the active functions, such as the number of instructions per clock cycle, and reorganizes the function-specific blocks among the levels of the hierarchy. Over time, this process tends toward block organizations that improve the performance metric.

Accelerated synchronization operations using fine grain dependency check are disclosed. A graphics multiprocessor includes a plurality of execution units and synchronization circuitry that is configured to determine availability of at least one execution unit. The synchronization circuitry to perform a fine grain dependency check of availability of dependent data or operands in shared local memory or cache when at least one execution unit is available.

Described herein is a system and method for flow state save/restore of a virtual filtering platform. A first instance of a driver manages policy and flow state for ongoing flows between client device(s) and virtual machine(s). The virtual filtering platform is transitioned from the first instance of a driver to a second instance of the driver by serializing the policy and state for the ongoing flows on the first instance of the driver using a one pass algorithm. The serialized policy and state for the ongoing flows can be de-serialized with the ongoing flows re-established and/or reconciled on the second instance of the driver in accordance with the de-serialized policy and state for the plurality of ongoing flows. In some embodiments, a memory management technique can use a single operating system memory allocation call to allocate memory for the transition, with the technique managing utilization of the allocation memory.

Embodiments presented herein techniques for balancing a multidimensional set of resources of different types within a distributed resources system. Each host computer providing the resources publishes a status on current resource usage by guest clients. Upon identifying a local imbalance, the host computer determines a source workload to migrate to or from the resources container to minimize the variance in resource usage. Additionally, when placing a new resource workload, the host computer selects a resources container that minimizes the variance to further balance resource usage.