Abstract: Some embodiments provide a hierarchical data service (HDS) that manages many resource clusters that are in a resource cluster hierarchy. In some embodiments, each resource cluster has its own cluster manager, and the cluster managers are in a cluster manager hierarchy that mimics the hierarchy of the resource clusters. In some embodiments, both the resource cluster hierarchy and the cluster manager hierarchy are tree structures, e.g., a directed acyclic graph (DAG) structure that has one root node with multiple other nodes in a hierarchy, with each other node having only one parent node and one or more possible child nodes.

Abstract: System and methods for generating a deployment, such as a software architecture, using existing telecommunication resources, such as microservices, data sources, and/or communication channels. A plain language message is received that describes requirements of a desired deployment. One or more entities are extracted from the plain language message. Based on the extracted entities, the system recommends one or more existing telecommunication resources for use in the desired deployment. In some implementations, recommendations are generated using a machine learning model that generates relevance scores for each of multiple existing telecommunication resources. A selection is received from among the recommended telecommunication resources, and the desired deployment is generated using the selected telecommunication resources.

Abstract: A method and system for automatic task time estimation and scheduling comprising the steps of: (1) storing a plurality of media items; (2) defining an aggregate task; (3) storing participant data and historical time data; (4) determining a plurality of metadata attributes; and (5) determining a final time estimate.

Abstract: Operating settings for a benchmarking operation may be determined, based at least in part, on one or more benchmarking task parameters. The operating settings may be used to establish a set of worker resources to send requests against a target until a stop condition is valid. Data may be collected from the target and used to train one or more machine learning systems in order to provide improved operating setting recommendations for subsequent benchmarking tasks.

Abstract: Computer-implemented methods, media, and systems for remediation of containerized workloads based on context breach at edge devices are disclosed. One example computer-implemented method includes monitoring telemetry data from a first software defined wide area network (SD-WAN) edge device, where the telemetry data includes multiple context elements at the first SD-WAN edge device. It is determined that a context change occurs for at least one of the context elements at the first SD-WAN edge device. It is determined that due to the context change, the first SD-WAN edge device does not satisfy one or more requirements for running one or more workloads scheduled to run. In response to the determination that the first SD-WAN edge device does not satisfy the one or more requirements, the at least one of the one or more workloads is offloaded from the first SD-WAN edge device to a second SD-WAN edge device.

Abstract: A network device may instantiate a first process in a first Multi-Access Edge Computing (MEC) cluster within a first MEC network; register the first MEC network at a MEC cloud; receive a call from the first process to obtain information from a second process in a second MEC cluster within a second MEC network that is registered at the MEC cloud; create a first message based on the call; send the first message from the first MEC cluster to the second MEC cluster; receive, at the first MEC cluster, a second message from the second MEC cluster, wherein the second message is provided by the second MEC cluster in response to receiving the first message; and extract a return value from the second message, wherein the return value is generated by the second process in response to receipt of the call in the second MEC cluster.

Abstract: Techniques for executing computing work by a plurality of chiplets are provided. The techniques include assigning workgroups of a kernel dispatch packet to the chiplets; by each chiplet, executing the workgroups assigned to that chiplet; for each chiplet, upon completion of all workgroups assigned to that chiplet for the kernel dispatch packet, notifying the other chiplets of such completion; and upon completion of all workgroups of the kernel dispatch packet, notifying a client of such completion and proceeding to a subsequent kernel dispatch packet.

Abstract: In accordance with an embodiment, described herein are systems and methods for providing tiered data storage in cloud infrastructure environments. A data storage service (block store) is adapted to automatically adjust the manner by which the data for a data volume or block volume (data/block volume), associated with a cloud instance, can be stored to meet the requirements of a performance tier. For example, responsive to selection of a particular performance tier, the storage of the data/block volume can be allocated between a first type of data storage associated with a first performance characteristics; and a second type of data storage associated with a second performance characteristics. A graphical user interface enables configuring data/block volumes to use particular performance tiers, and/or to support automatic tuning.

Abstract: The present disclosure provides a parallel decision system and method for distributed data processing. The system includes: an initial logical node generation assembly, a logical node traversal assembly, a predetermined configuration cost computation assembly, and a parallel decision assembly. The initial logical node generation assembly is configured to receive task configuration data input by a user to generate an initial logical node topology for the distributed data processing system. The logical node traversal assembly is configured to traverse the initial logical node topology to obtain a predetermined configuration in the initial logical node topology. The predetermined configuration cost computation assembly is configured to compute a transmission cost of each predetermined configuration and a cost sum.

Abstract: A system for adapting an image noise removal model based on a device processing capability receives, from a computing device, a request to adapt an image noise removal module for the computing device. The system compares a processing capability of the computing device with a threshold processing capability. The system determines whether the processing capability is greater or smaller than the threshold processing capability. In response to determining that the processing capability is greater than the threshold processing capability, the system sends a version of the image noise removal module that is adapted for computing devices with processing capabilities less than the threshold processing capability, where the version of the image noise removal module is adapted to have a number of neural network layers less than a threshold number of neural network layers.

Abstract: A method and system for source versus target metadata-based data integrity checking. Concerning backup operations directed to protecting given data, said given data may be subjected to corruption detection at the source prior to initiating a backup operation, however, said given data may not be checked for data integrity following transfer of said given data to a target storage medium prior to committing the said given data thereto. That is, at least presently, the prospect of data corruption compromising given data during the time window through which the given data journeys, usually via a network, from its source to a target storage medium, is often overlooked. The disclosed method and system, accordingly, propose a scheme directed to detecting corruption amongst data transferred from a source to a target storage medium, and handling said data given the determined integrity of said data.

Abstract: Systems and methods provide a provisioning framework for a distributed graphics processing unit (GPU) service. A network device in a network receives, from an application, a service request for multi-access edge compute (MEC)-based virtual graphic processing unit (vGPU) services. The network device receives real-time utilization data from multiple MEC clusters in different MEC network locations and generates a utilization view of the multiple MEC clusters in the different MEC network locations. The network device selects, based on the real-time utilization view, one of the different MEC network locations to provide the vGPU services and instructs a of the multiple MEC clusters in the one of the different MEC network locations to perform container provisioning and service provisioning for the vGPU services.

Abstract: A method for task offloading based on power control and resource allocation in the Industrial Internet of Things includes establishing a computing model for computation tasks at different offloading locations, constructing communication power control, resource allocation and computation offloading problems as a mixed integer non-linear programming model, solving them using a deep reinforcement learning algorithm to obtain an optimal strategy for offloading of the computation tasks, thus achieving communication power optimization and cross-domain resource allocation.

Abstract: A method for performing a parallelized heapsort operation may include updating, by a first worker thread, a first buffer while a second worker thread updates a second buffer in parallel. The first worker thread may update the first buffer by adding, to the first buffer, elements from a first partition of a dataset. The second worker thread may update the second buffer by adding, to the second buffer, elements from a second partition of the dataset. Upon the first buffer reaching a threshold size, the first worker thread may acquire a lock for the first worker thread to update a heap based on the first buffer while the second worker thread is prevented from updating the heap based on the second buffer. A result of a top k query comprising a k quantity of smallest elements from the dataset may be generated based on the heap.

Abstract: An apparatus with a solid state drive (SSD) configured to manage storage resources for proof of space activities. The SSD has a host interface configured to receive at least read commands and write commands from an external host system. The SSD has memory cells formed on at least one integrated circuit die, and a processing device configured to control executions of the read commands and the write commands. In response to an indication of a storage space request for the host system, the apparatus identifies an application using a proof of space plot stored in a portion of the solid state drive, requests the application to separate from the proof of space plot, and then delete a namespace in which the proof of space plot is stored to release storage resources occupied by the proof of space plot to meet the storage space request.

Abstract: Load balancing VM selection and movement is disclosed, including: selecting a combined predicted metric to be improved with respect to a plurality of storage devices; and determining a set of VMs to be moved from a first storage device included in the plurality of storage devices to a second storage device included in the plurality of storage devices such that the selected combined metric to improve is predicted to improve.

Abstract: A server that includes a graphics processing unit (GPU) may receive, from a first application that is remote from the server, a first request to reserve a first number of cores of the GPU for a first amount of time. The server may also receive, from a second application that is also remote from the server, a second request to reserve a second number of cores of the GPU for a second amount of time that at least partly overlaps the first amount of time. The server may determine that the first request is associated with a higher priority than the second request and, in response, may reserve the first number of cores for the first amount of time for the first application. The server may send, to the first application, an indication that the first number of cores have been reserved as requested by the first application.

Abstract: A processor includes a front end, an execution unit, a retirement stage, a counter, and a performance monitoring unit. The front end includes logic to receive an event instruction to enable supervision of a front end event that will delay execution of instructions. The execution unit includes logic to set a register with parameters for supervision of the front end event. The front end further includes logic to receive a candidate instruction and match the candidate instruction to the front end event. The counter includes logic to generate the front end event upon retirement of the candidate instruction.

Abstract: Methods and apparatus are disclosed that deploy a hybrid workload domain. An example apparatus includes a resource discoverer to determine whether a first bare metal server is available and a resource allocator to allocate virtual servers for a virtual server pool based on an availability of the virtual servers and, when the first bare metal server is available, allocate the first bare metal server for a bare metal server pool. The example apparatus further includes a hybrid workload domain generator to generate, for display in a user interface, a combination of the virtual server pool and the bare metal server pool and generate a hybrid workload domain used to run a user application based on a user selection in a user interface, the hybrid workload domain including virtual servers from the virtual server pool and bare metal servers from the bare metal server pool.

Abstract: To provide a media data processing system that can suppress a decrease in a request processing rate while suppressing an increase in response time in media data recognition processing where it is difficult to properly estimate the load. A first load estimation unit 5 estimates range of processing load of media data recognition processing based on header information of media data. A determination unit 31 determines whether to allow or disallow execution of the media data recognition processing, or to estimate the processing load, based on the range of the processing load. A second load estimation unit 6 estimates the processing load of the media data recognition processing based on content of the media data when it is determined to estimate the processing load. The determination unit 31 determines whether to allow or disallow the execution of the media data recognition processing based on the processing load.

Abstract: A data processing method includes receiving a request to perform a calculation, identifying, based on the request, data items needed to perform the calculation and retrieving the data items from a data store, storing, in memory, the items, generating graphs for the calculation, wherein each graph comprises one or more nodes, each node comprising instructions to perform at least a portion of the calculation and at least one data item needed by the portion of the calculation, executing each of the graphs to generate a result for the calculation by traversing the graph and processing each node using the instructions of the node and the at least one data item of the node, wherein executing is performed without accessing the data store, and storing, in the data store, the result.

Abstract: Methods, apparatus, systems, and articles of manufacture are disclosed for Strategy-Driven Optimization of Computer Resource Configurations in a Cloud Environment. Disclosed examples include a non-transitory computer readable storage medium comprising instructions that, when executed, cause processor circuitry to: monitor consumption of cloud resources associated with a containerized workload; associate the consumption with a monetary cost; and generate a notification to notify a user the cost exceeds the threshold, the notification including workload metrics associated with the monetary cost.

Abstract: Virtual machine environments are provided in the switches that form a network, with the virtual machines executing network services previously performed by dedicated appliances. The virtual machines can be executed on a single multi-core processor in combination with normal switch functions or on dedicated services processor boards. Packet processors analyze incoming packets and add a services tag containing services entries to any packets. Each switch reviews the services tag and performs any network services resident on that switch. This allows services to be deployed at the optimal locations in the network. The network services may be deployed by use of drag and drop operations. A topology view is presented, along with network services that may be deployed. Services may be selected and dragged to a single switch or multiple switches. The management tool deploys the network services software, with virtual machines being instantiated on the switches as needed.

Abstract: A method and system for analyzing and connecting computer-based actions into sentences may include for a series of computer-based actions, determining the case ID for the action for each action where an identifier or case ID can be determined, creating sequences of subsets of the series of computer-based actions using the case ID, and merging sequences having computer-based actions having the same case ID. A set of case IDs may be extracted from the actions using a clustering algorithm based on features of potential case IDs such as gaps in appearance of potential case IDs in a sequence of actions and consecutive appearances of potential case IDs in a sequence of actions. The extracted case IDs may be used when creating sequences.

Abstract: A system is a system for managing release of a developed program to a production environment, and includes a plurality of servers in which a blockchain is installed, a means for detecting an event related to release of a program, and a means for generating a block of the blockchain, which contains uniqueness information that assures uniqueness of the program when the event was detected, and specification information that specifies the detected event.

Abstract: An example method to manage post provisioning operations of a virtual computing instance in a heterogeneous cloud environment is disclosed. The virtual computing instance may be provisioned by a first management entity and configured to receive a command from a second management entity. The method includes defining the instance with a dynamic type by the first management entity and repeatedly finding the dynamic type with one or more finder workflows to determine whether the virtual computing instance is terminated based on the command from the second management entity. In response to not finding the dynamic type within the heterogeneous cloud environment, the method further includes creating a catalog item for the virtual computing instance in a common service catalog and managing one or more resources allocated for the virtual computing instance based on the created catalog item.

Abstract: A disclosed method for managing a multi-node, HCI cluster provisions a central manager of the HCI cluster with an event listening service. The service is then registered with an agent in each HCI node. The service, when triggered by a management network configuration event, causes the agent to modify a management network configuration of the node in accordance with the event. Disclosed methods may respond to successfully applying a management network configuration change by sending a success notification to the central manager and awaiting a confirmation from the central manager. If confirmation is not received within a timeout interval, a timeout occurs and the management network configuration is rolled back to a last known good configuration. If the configuration change cannot be successfully applied, a failure notification is sent to the central manager and the configuration of the node is rolled back to a last known good configuration.

Abstract: A method, apparatus, system, and computer program product for configuring a computing environment. A configuration profile is identified by a computer system for the computing environment that is to be deployed in which the computing environment meets a security policy to run an application in the computing environment. A determination is made, by the computer system, as to whether the configuration profile for the computing environment meets the security policy for running the application in the computing environment. The configuration profile for the computing environment is deployed, by the computer system, to configure the computing environment for the application in response to the configuration profile meeting the security policy.

Abstract: A scheduler manages execution of a plurality of data-collection jobs, assigns individual jobs to specific forwarders in a set of forwarders, and generates and transmits tokens (e.g., pairs of data-collection tasks and target sources) to assigned forwarders. The forwarder uses the tokens, along with stored information applicable across jobs, to collect data from the target source and forward it onto an indexer for processing. For example, the indexer can then break a data stream into discrete events, extract a timestamp from each event and index (e.g., store) the event based on the timestamp. The scheduler can monitor forwarders' job performance, such that it can use the performance to influence subsequent job assignments. Thus, data-collection jobs can be efficiently assigned to and executed by a group of forwarders, where the group can potentially be diverse and dynamic in size.

Abstract: System resource utilizations by one or more core functionalities of a data storage system is monitored. A system resource allocation for one or more applications of the data storage system is determined based at least in part on the monitored system resource utilizations. A number of virtualization resource units allocated to at least one of the one or more applications is scaled based at least in part on the determined system resource allocation. The number of virtualization resource units allocated to the at least one of the one or more applications is based on a corresponding ratio associated with the at least one of the one or more applications.

Abstract: Methods, apparatus, systems, and articles of manufacture are disclosed to intelligently select content items for provisioning. In one example, the apparatus includes content library querying circuitry that queries a subscription guide database table to generate a list of subscriber content items linked to a target published content item indicated in the request. The apparatus also includes content library selection circuitry to select a first content item from a group of content items that include the generated list of the subscriber content items and the target published content item, the first content item corresponds to a first data access cost that is less than or equal to a second data access cost of a second content item in the group. The apparatus also includes virtual machine provisioning circuitry to provision the virtual machine on the target host with access to the first content item.

Abstract: A method of collecting and reporting inventory of resources deployed in a data center that includes hardware resources, a virtualization management software executed to provision virtual resources from the hardware resources, and a cloud management server executed to provision the virtual resources for tenants of the data center, includes the steps of: executing a first API call to the virtualization management software to collect first inventory of virtual resources deployed in the data center and a second API call to the cloud computing management software to collect second inventory of virtual resources deployed in the data center; storing the first and second inventory; and in response to an inventory request from a central orchestrator, initially sending a subset of the stored first and second inventory to the central orchestrator in accordance with parameters included in the inventory request, and thereafter sending updates to the subset to the central orchestrator periodically.

Abstract: Various embodiments of the present disclosure are directed to improved systems, apparatuses, methods, and computer program products for managing and synchronizing independent computing resources. In this regard, embodiments may manage and/or synchronize computing resources, data, and the like between the embodiments and various independent third-party systems, while handling transmission errors that occur during synchronization. In an example context, embodiments manage synchronization of subscriber profiles between the various systems. Some embodiments manage computing resources for processing one or more unprocessed actions of various types, for example by scaling executed process instances based on an action queue length and a queue length minimum threshold and/or queue length maximum threshold.

Abstract: Performance optimization is achieved by clarifying cache usage characteristics of each application from usage conditions of physical resources (caches) in real time and automatically controlling the cache usage amount of each application. Thus, a system includes a main memory to and from which data is written and read, a level 3 cache memory which can be accessed faster than the main memory, a CPU core configured to execute processing by performing write and read to and from the memory and the cache, a usage amount measurement unit configured to measure a usage condition of a cache of each virtual machine (13a to 13c) executed by the CPU core, an allocation amount calculation unit configured to calculate cache capacity to be allocated to each virtual machine (13a to 13c) from the usage condition, and a control unit configured to allocate the cache capacity to each virtual machine (13a to 13c).

Abstract: This disclosure describes a storage aggregator controller with metadata computation control. The storage aggregator controller communicates, via a host interface, over a computer network with one or more remote hosts, and also communicates, via a storage device interface, with a plurality of local storage devices, which are separate from the remote host(s) and which have respective non-volatile memories. The storage aggregator controller manages the local storage devices for storage or retrieval of media objects. The storage aggregator controller also governs a selective computation, at aggregator control circuitry or at a storage device controller of one or more of the storage devices, of metadata that defines content characteristics of the media objects that are retrieved from the plurality of storage devices or that are received from the one or more hosts over the computer network for storage in the plurality of storage devices.

Abstract: Some embodiments provide a non-transitory machine-readable medium that stores a program. The program receives data from a subset of a first set of exchange tables stored in a first database of a source system. The program further creates a second set of exchange tables in a second database of the device. The program also creates a set of shadow tables in the second database of the device. The program further adapts the second set of exchange tables and the set of shadow tables to a higher upgrade version. The program also performs object generation based on the second set of exchange tables and the set of shadow tables. The program further drops the set of shadow tables. The program also receives a set of application tables from the source system. The program further adapts the set of application tables to the higher upgrade version.

Abstract: A computer-implemented method to estimate an execution time for large query batches. The method includes identifying a batch of queries, where the batch of queries where each query is configured to retrieve data from a database. Selecting a first query and generating an access path for the first query including one or more mini plans. Building a learning model for the one or more mini plans. Estimating an estimated execution time for each of the one or more mini plans. Adding each of the estimated execution times for each of the one or more mini plans to determine a first total estimated time for the first query. Displaying the first total estimated time.

Abstract: Methods, systems, and computer storage media are provided for providing an application programming interface (“API”) for executing a query are provided. The API call, scheduling, and assignment processing engines may generate and schedule assignments, along with associated dataset portions, based on the query. Based on the assignments and datasets, the assignments may be scheduled for execution by processes of nodes in a distributed system. The API call and query may also include implementation instructions that require storing dataset portions into memory that is associated with a process executing the assignment for an in-memory database implementation. As the assignment is executed using a dataset portion that is stored on memory associated with a process, results may be stored in a file. Once each assignment has been executed and has a result file, the result files are merged to provide results of the query and API call to the user.

Abstract: The present disclosure is directed to a novel system for using a distributed register to generate, manage, and store data for interest-pooled time deposit resource accounts. The invention leverages a pooled resource account approach, allowing for multiple disparate resource accounts to benefit from an enhanced interest return by pooling resource accounts. The system components of the invention contemplate the use of distributed register technology to provide a verified ledger of information related to one or more resource accounts, as well as store system data, user data, and metadata related to the movement and management of resources. By using a distributed register approach to store and verify data related to time-dependent resource account services, the invention provides an automated system and methods for enhancing the flow of sensitive verified information, reducing the need for manual review and increasing the speed at which various resource account services can be validated and executed.

Abstract: Disaster recovery operations are disclosed. A copy of data may include segments and delta journals or journals. RTO can be reduced by pre-processing the delta journals after the segments and delta journals are uploaded. The RTO is improved during recovery because the need to process the delta journals is eliminated. Applying the delta journals can be performed to both reduce cost and RTO using graphs.

Abstract: An image forming apparatus to communicate with an information processing apparatus includes a reception unit and a transmission unit. The reception unit receives information for identifying a print service included in the information processing apparatus. The transmission unit transmits, to the information processing apparatus, a prohibition condition. The prohibition condition represents a condition relating to a print setting that cannot be set depending on capabilities of a printing apparatus and of the print setting. The prohibition condition is for a print service corresponding to the information received by the reception unit.

Abstract: In some embodiments, infrastructure data and service data is received for a computing infrastructure. The infrastructure data indicates resources in the computing infrastructure, and the service data indicates services to be orchestrated across the computing infrastructure. An infrastructure capacity model is generated, which indicates a capacity of the computing infrastructure over a particular time window. Service-to-resource placement options are also identified, which indicate possible placements of the services across the resources over the particular time window. Resource inventory data is obtained, which indicates an inventory of resources that are available to add to the computing infrastructure during the particular time window. An infrastructure capacity plan is then generated, which indicates resource capacity allocation options over the time slots of the particular time window. Resource capacities for the services are then allocated in the computing infrastructure.

Abstract: A virtual device management system for creating and implementing virtual devices and associated methods are provided herein. The system presents to a user a list of devices available for combining to create the virtual device. When the system receives from the user a selection of two or more devices that the user wishes to combine to create the virtual device, the system presents to the user a list of available functions that can be performed by the virtual device based on the selected devices. The system then receives from the user a selection of one or more functions and configures the virtual device to perform the selected functions. The system can also be configured to present, via a user interface via a mobile application running on a user device, a plurality of virtual devices with predetermined functionalities that are created by detecting the various connected devices within an IoT network.

Abstract: A network-based data store maintaining storage volumes for file systems at storage clients may implement dynamic storage volume scaling. A scaling event may be detected according to a specified scaling policy for a storage volume maintained at a network-based data store that is mounted at a storage client and configured for a file system. In a least some embodiments, the scaling policy for the storage volume may be received from the storage client. In response to detecting the storage event, storage maintaining the storage volume may be modified according to the scaling policy to scale the storage volume. An indication of the scaling event may be sent to the storage client in order to update the file system at the storage client so that the file system may utilize the scaled storage volume.

Abstract: Systems and methods for managing computer block storage for a computer application include calculating an optimal required block storage capacity based on the storage needs of the application; provisioning block storage of the optimal capacity; receiving at least one block storage usage metric of the application; using a machine learning based model, trained on historic data of at least one application, to identify at least one future time at which a block storage capacity adjustment is required; and adjusting the block storage capacity within a time of the future time at which the block storage capacity adjustment is required.

Abstract: A task scheduling system that can be used to improve task assignment for multiple satellites, and thereby improve resource allocation in the execution of a task. In some implementations, configuration data for one or more satellites is obtained. Multiple objectives corresponding to a task to be performed using the satellites, and resource parameters associated with executing the task to be performed using the satellites are identified. A score for each objective included in the multiple objectives is computed by the terrestrial scheduler based on the resource parameters and the configuration data for the one or more satellites. The multiple objectives are assigned to one or more of the satellites. Instructions are provided to the one or more satellites that cause the one or more satellites to execute the task according to the assignment of the objectives to the one or more satellites.

Abstract: A device may include a processor configured to determine latency budgets for particular microservices for an application associated with a user equipment (UE) device, wherein the microservices are deployed in a cloud computing center. The processor may be further configured to determine that a measured latency for a particular microservice has exceeded a latency budget for the particular microservice by at least a latency budget threshold; deploy the particular microservice on a Multi-Access Edge Computing (MEC) network associated with a base station servicing the UE device, based on determining that the measured latency for the particular microservice has exceeded the latency budget for the particular microservice by at least the latency budget threshold; and send a recommendation to the UE device to use the particular microservice deployed at the MEC network.

Abstract: Methods, computer systems, and apparatus, including computer programs encoded on computer storage media, for allocating computing resources for a data object. The system obtains a plurality of characteristics of a data object, and estimates, from the obtained characteristics, one or more cumulative sustainability metrics characterizing one or more categories of energy consumption during a life-cycle of the data object. The system further determines, from the cumulative sustainability metrics, allocations of one or more computing resources to the data object to optimize one or more objectives including minimizing a cumulative carbon cost during the life-cycle of the data object.

Abstract: A method of distributed resource management in a distributed computing system includes determining usage of respective hardware resources by an application and generating usage metrics for the application, and assigning the application to a cluster of hardware resources to optimize diversity of usage of hardware resources in the cluster and to enhance utilization of the hardware resources by applications running in that cluster. The diversity of usage of the hardware resources is determined from respective usage metrics of the respective applications running in that cluster. The diversity of usage of the hardware resources in the cluster is optimized by assigning the application to a diversity pool of hardware resources adapted to minimize interference when applications assigned to the diversity pool of hardware resources access the hardware resources in the diversity pool and assigning applications from different diversity pools to the cluster of hardware resources.

Abstract: Methods and apparatus to execute a workload in an edge environment are disclosed. An example apparatus includes a node scheduler to accept a task from a workload scheduler, the task including a description of a workload and tokens, a workload executor to execute the workload, the node scheduler to access a result of execution of the workload and provide the result to the workload scheduler, and a controller to access the tokens and distribute at least one of the tokens to at least one provider, the provider to provide a resource to the apparatus to execute the workload.