Provided is an operation method of a NUMA system, which includes: designating a page scan range including a plurality of pages; identifying a detour value for each of the plurality of pages; determining whether a detour value of a current target scan page is the same as the reference detour value; and releasing a connection of the current target scan page from the page table when determining that the detour value of the current target scan page is the same as the reference detour value.

Techniques for executing machine learning (ML) models including receiving an indication to run an ML model on a processing core; receiving a static memory allocation for running the ML model on the processing core; determining that a layer of the ML model uses more memory than the static memory allocated; transmitting, to a shared memory, a memory request for blocks of the shared memory; receiving an allocation of the requested blocks; running the layer of the ML model using the static memory and the range of memory addresses; and outputting results of running the layer of the ML model.

For a first virtual machine (VM) executing in a physical machine, a second VM is instantiated in the physical machine, the first VM using a physical adapter installed in the physical machine, the first VM virtualizing a portion of physical memory of the physical machine, the second VM virtualizing the physical adapter. The second VM is deployed using a memory mapping virtualizing the portion of physical memory. Checkpointing of an application executing in the first VM is caused, generating application state data of the application. The application is caused to execute in the second VM using the application state data. Process data of the application is caused to be updated in the second VM, the updating instructing the application to use the memory mapping.

Disclosed are various embodiments for optimizing hypervisor paging. A hypervisor can save a machine page to a swap device, the machine page comprising data for a physical page of a virtual machine allocated to a virtual page for a process executing within the virtual machine. The hypervisor can then catch a page fault for a subsequent access of the machine page by the virtual machine. Next, the hypervisor can determine that the physical page is currently unallocated by the virtual machine in response to the page fault. Subsequently, the hypervisor can send a command to the swap device to discard the machine page saved to the swap device in response to a determination that the physical page is currently unallocated by the virtual machine.

Systems and corresponding methods employ an object-oriented (OO) memory (OOM) to effect inter-hardware-client (IHC) communication among a plurality of hardware clients included in same. A system comprises a centralized OOM and the plurality of hardware clients communicate, directly, to the centralized OOM device via OO message transactions. The centralized OOM device effects IHC communication among the plurality of hardware clients based on the OO message transactions. Another system comprises a plurality of OO memories (OOMs) capable of inter-object-oriented-memory-device communication. A hardware client communicates, directly, to a respective OOM device via OO message transactions. The inter-object-oriented-memory-device communication effects IHC communication among the plurality of hardware clients based on the OO message transactions.

A resource management system of an application takes various actions to improve or maintain the health of the application (e.g., keep the application from becoming sluggish). The resource management system maintains a reinforcement learning model indicating which actions the resource management system is to take for various different states of the application. The resource management system performs multiple iterations of a process of identifying a current state of the application, determining an action to take to manage resources for the application, and taking the determined action. In each iteration, the resource management system determines the result of the action taken in the previous iteration and updates the reinforcement learning model so that the reinforcement learning model learns which actions improve the health of the application and which actions do not improve the health of the application.

Power consumption can be reduced by selective memory chip hibernation. For example, a computing device can allocate first data associated with a first processing operation of a user device to a first chip of a dynamic random access memory (DRAM) of the user device. The computing device can allocate second data associated with a second processing operation of the user device to a second chip of the DRAM of the user device. The computing device can determine the first processing operation has been inactive for a predetermined period of time and migrate the first data from the first chip of the DRAM to a storage device of the user device. The computing device can hibernate the first chip of the DRAM while maintaining power to the second chip of the DRAM for continuing to perform the second processing operation.

Systems and methods are provided for intercepting computing requests and modifying the execution timing thereof based on thresholds and minimum performance criteria and/or adjusting hosted services plans in order to monitor and control costs of hosting software applications on hosted provider computing resources.

Shared resource interference detection techniques are described. In an example, a resource detection module supports techniques to quantify levels of interference through use of working set sizes. The resource detection module selects working set sizes. The resource detection module then initiates execution of code that utilizes the shared resource based on the first working set size. The resource detection module detects a resource consumption amount based on the execution of the code. The resource detection module then determines whether the detected resource consumption amount corresponds to the defined resource consumption amount for the selected working set size.

A protection system is provided for delivering runtime security to a task including a workload container. The protection system uses a sidecar to limit access of the workload container to a standard library of the operating system running the workload container by modifying the task so that the sidecar is executed before the workload container. The sidecar places a guard loader into a shared volume and binds the workload container, such that calls to the workload container are passed to an agent binary. The agent binary compares requested calls from the workload container to a policy to approve and/or deny the requested calls. If the requested call is approved, then the requested call is passed to the standard library.