Abstract: Methods, systems, and computer-readable storage media for executing, within the container orchestration system, the application using one or more resource units, each resource unit including an application container and an ancillary container, the ancillary container executing a set of GPU metric exporters, receiving, from the application and for each resource unit, a GPU metric including a GPU utilization associated with a respective resource unit, determining, for each resource unit, a custom GPU metric based on a respective GPU metric, the custom GPU metric aggregating values of the respective GPU metric over a time period, determining, by an autoscaler, an average GPU metric based on one or more custom GPU metrics, and selectively scaling, by the autoscaler, the application within the container orchestration system based on the average GPU metric by adjusting a number of resource units executing the application.

Abstract: 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.

Abstract: Systems, methods, and apparatuses for resource bandwidth monitoring and control are described. For example, in some embodiments, an apparatus comprising a requestor device to send a credit based request, a receiver device to receive and consume the credit based request, and a delay element in a return path between the requestor and receiver devices, the delay element to delay a credit based response from the receiver to the requestor are detailed.

Abstract: Resegmenting chunks of data for load balancing is disclosed. A plurality of first chunks of data is received. The plurality of first chunks of data includes one or more entries that include raw data produced by a component of an information technology environment and that reflects activity in the information technology environment. The plurality of first chunks of data is resegmented into a plurality of second chunks of data based on a source type of the plurality of first chunks. A first subset of the plurality of second chunks of data is distributed to a first indexer of a set of indexers. An occurrence of a trigger event is determined, and in response to the trigger event, a second subset of the plurality of second chunks of data is distributed to a second indexer of the set of indexers.

Abstract: Decentralized cloud service assessment includes using a self-executing data structure, an error confirmation capsule (ECC) generated in response to a cloud service failure experienced by a cloud service client (CSC). One or more technical performance indicia corresponding to the cloud service failure are extracted from the ECC in response to the validating. The one or more technical performance indicia are compared to one or more electronically stored predefined performance norms of a cloud service provider (CSP) associated with the cloud service failure. Based on the comparing, a comparative ranking of the CSP is determined. A graphical user interface display is generated based on comparative rankings of the CSP and one or more other CSPs.

Abstract: A method to deploy a plurality of event-driven application components of an event-driven application in a distributed computing environment is described. The method includes automatically analyzing application source code of the event-driven application, using one or more processors, to identify relationships between the plurality of event-driven application components. Thereafter, a set of rules are applied to, based on the automatic analysis, generate assignment data recording assignments of event-driven application components to a plurality of computational nodes in the distributed computing environment. The set of rules is also applied to determine component requirements for each of the plurality of event-driven application components required to support execution at an assigned computational node in the distributed computing environment.

Abstract: Described embodiments provide systems and method for determining action insights to address, correct or fix application failures. A device can identify an assignment of a user to a virtual machine and the assignment can enable launch of an application with use of the same virtual machine on behalf of the user. The virtual machine can be one of a delivery group of virtual machines. A threshold can be determined for the virtual machine that indicates a likely failure of the virtual machine to launch the application based data about performance of the delivery group. The device can modify the assignment of a user from the virtual machine to another virtual machine of the delivery group based on a comparison of a load on the virtual machine and the determined threshold, so as to avoid failure of the application to launch.

Abstract: An information processing apparatus calculates a first load value indicating the load of a plurality of first jobs whose used node counts are less than or equal to a threshold among jobs waiting for execution, and a second load value indicating the load of a plurality of second jobs whose used node counts exceed the threshold among the waiting jobs. If load conditions including one where the first load value exceeds the second load value are satisfied, the information processing apparatus combines two or more first jobs among the plurality of first jobs to create a combined job whose used node count exceeds the threshold. The information processing apparatus causes a first node group to execute the remaining first jobs among the plurality of first jobs and causes a second node group to execute the combined job and the plurality of second job.

Abstract: A resource scheduling method and system, and a computer-readable storage medium are provided. The method comprises: a control instruction inputted by a user (S101) is acquired; on the basis of the control instruction, an intended device (S102) is determined; in the case where the intended device is not able to be started, with regard to each function of the intended device, a constructed knowledge graph is queried for alternative devices associated with the function, the knowledge graph having a plurality of functions stored therein as well as devices associated with the plurality of functions (S103, S105); and a device to be started is determined from all the alternative devices and starting the determined device (S106). In the resource scheduling method, in the case where an intended device specified by a user is not able to be started, alternative devices are appropriately determined.

Abstract: A processing system monitors and synchronizes parallel execution of workgroups (WGs). One or more of the WGs perform (e.g., periodically or in response to a trigger such as an indication of oversubscription) a waiting atomic instruction. In response to a comparison between an atomic value produced as a result of the waiting atomic instruction and an expected value, WGs that fail to produce a correct atomic value are identified as being in a waiting state (e.g., waiting for a synchronization variable). Execution of WGs in the waiting state is prevented (e.g., by a context switch) until corresponding synchronization variables are released.

Abstract: System and methods providing for categorizing individual virtual machines, as well as the associated application that they form by working in concert, into groups based on the feasibility of hosting the processes that occur on a virtual machine within a container, as well as the relative difficulty of doing so on a virtual machine and application level. The data used to create these scores is collected from the individual machines, at regular intervals through the use of an automated scoring engine that collects and aggregates the data. Said data is then analyzed by the system, that with the aid of passed in configuration data, is configured to generate the scores to allows for an educated and focused effort to migrate from hosting applications on virtual machines to hosting applications on containers.

Abstract: One or more computer operations are triggered and executed when multiple sets of conditions that define a split trigger and that are evaluated at multiple locations are satisfied. The operations may involve allocating or deallocating computing resources (e.g., by establishing or tearing down a persistent connection), generating a message to be displayed on a computing device, altering stored data, and/or other activity. In an illustrative client/server computing environment, a first set of client conditions of a first split trigger is evaluated at a client device, using parameters observable at the client without interaction with a server. When the first of client conditions is satisfied, the client transmits a signal to a server that then evaluates a first set of server conditions of the first split trigger. If the first sets of client conditions and server conditions are satisfied, the computer operation(s) associated with the split trigger are executed.

Abstract: An apparatus comprises a processing device configured to identify a plurality of workloads to be scheduled for execution on a storage system and to analyze the plurality of workloads to predict a magnitude of storage capacity savings achieved by applying one or more deduplication algorithms to data of the plurality of workloads. The processing device is further configured to determine a prioritization of the plurality of workloads based at least in part on the predicted magnitude of the storage capacity savings achieved by applying the one or more deduplication algorithms to the data of the plurality of workloads, and to schedule the plurality of workloads for execution on the storage system based at least in part on the determined prioritization of the plurality of workloads.

Abstract: This application provides a method and an apparatus, for accelerating cold startup of an application. The method includes after identifying an event that instructs an operating system of a terminal to cold start up an application, obtaining, from a plurality of dimensions, current status information related to the cold startup of the application, where the current status information includes a hardware configuration of the terminal, current load of the operating system of the terminal, resource overheads for cold starting up the application, and duration corresponding to each of a plurality of tasks in a process of cold starting up the application. The method also includes determining, by analyzing the current status information, a plurality of objects that need to be optimized in the current process of cold starting up the application; and then obtaining, based on the determined objects.

Abstract: Systems and methods for live updating virtualization systems are disclosed. One method may include identifying a container on a first virtual machine of a virtualization system. The container may include at least one process associated with performing a task on a first virtual machine. The method may also include determining a first communication path between the process included in the container on the first virtual machine and a hypervisor of the virtualization system, and migrating the container and the process included in the container to a second virtual machine. The second virtual machine may be distinct from the first virtual machine. Additionally, the method may include determining a second communication path between the process included in the migrated container on the second virtual machine and the hypervisor of the virtualization system.

Abstract: An example method includes receiving historical usage data associated with computing services provided by distributed servers and an inference model. The inference model is configured to receive a request and make an inference based on the request. The method further includes training a machine learning model to determine a correlation between usage of a first computing service of the and usage of the inference model. The correlation indicates that a first spike in usage of the first computing service precedes a second spike in usage of the inference model. The method further includes receiving, in real-time, current usage data associated with the first computing service. The method further includes determining, based on the current usage data and the correlation, that the current usage data is indicative of the first spike in usage of the first computing service that precedes the second spike in usage of the inference model.

Abstract: Executing geospatial data workflows includes receiving a representation of a first workflow. It further includes receiving a request to execute a second workflow, the second workflow including a reference to the first workflow. At least one of the first workflow and the second workflow comprises at least one geospatial transformation operation. It further includes determining that the first workflow referenced by the second workflow is permitted to be accessed. It further includes, based at least in part on determining that the first workflow is permitted to be accessed, executing the second workflow, including executing the representation of the first workflow comprising the at least one geospatial transformation operation.

Abstract: Techniques for process execution trend prediction and visualization are disclosed. The disclosed system receives a process execution request to be executed on a set of targets. The request may include request characteristics, such as a request type and computations to be performed during execution. The system analyzes the request characteristics to determine the computations to execute and for initiates request execution on the targets. Based on the analysis, the system generates predictions regarding the execution, including an estimated completion time. During execution, the system displays various attributes of the execution in a dynamically updating visualization. The system also provides real-time recommendations on how the process can be optimized, such as to reduce execution time and errors.

Abstract: Described are examples for tracking memory usage of a driver. A memory allocation request related to the driver to allocate a portion of memory for the driver can be traced in a kernel mode of an operating system. One or more associated allocation parameters can be recorded, and an allocation history of the driver over a period of time can be reported during execution of the driver and based on the one or more allocation parameters indicated by the memory allocation request.

Abstract: A method may include receiving, via a first computing node of a cluster of computing nodes in a container orchestration system, a pod from a second computing node in the cluster of computing nodes. The method may also involve retrieving an image file that includes containers from a registry, such that the pod detail an indication of a location of the image file in the registry. The method then involves generating a package based on the one or more containers and one or more mapped commands implementable by a control system in an operational technology (OT) domain. The mapped commands correspond to operations performable by the control system that corresponds to commands specified in the containers. The method may then involve storing the package in a filesystem shared with the control system.