This disclosure describes dynamically placing workloads using cloud service energy efficiency. The techniques include obtaining energy efficiency metrics (EEMs) that indicate the carbon footprint for different data centers of cloud service providers. In some configurations, an Energy Efficiency Quotient (EEQ) may be generated by an Energy Telemetry Engine (ETE) that indicates the energy efficiency for each data center/Point of Presence (POP) where a workload may be migrated/hosted. The ETE can be used to rank the different host locations (e.g., different data according to their EEQ. In some examples, one or more other metrics (e.g., latency, bandwidth, . . . ) may be used to identify any POPs that do not meet specified conditions (e.g., latency constraints, bandwidth constraints, . . . ). When a suitable host location is determined (e.g. a POP meets both the performance and EEQ specifications), the workload may be placed onto one or more resources of the selected data center.

Some embodiments provide a method for providing access in a scalable manner to resources in a first datacenter to clients operating in one or more public clouds. The method of some embodiments implements with multiple machines a public-cloud proxy to connect clients in the public cloud(s) to a reverse proxy in the first datacenter. For instance, in response to a request to access a first resource in the first datacenter from a first client executing outside of the first datacenter, the method: (1) assigns a first proxy-implementing machine operating outside of the first datacenter to establish a first connection with the first client, (2) assigns a second proxy-implementing machine operating outside of the first datacenter to establish a second connection with the reverse proxy that operates in the first datacenter and that provides access to the first resource, and (3) establishes a third connection between the first and second proxy-implementing machines to forward messages between the first client and the reverse proxy through the first, second, and third connections.

An example method of reserving a resource of virtualized infrastructure in a data center on behalf of a client includes: obtaining, by a resource lock manager from a topology manager, a sub-topology for the resource from a resource topology of the virtualized infrastructure; setting, by the resource lock manager, an exclusive lock on the resource and on each of at least one descendant in the sub-topology for the resource, each exclusive lock disallowing any other lock on its respective resource; setting, by the resource lock manager, a limited lock on each ancestor in the sub-topology for the resource, each limited lock allowing any other limited lock on its respective resource; and notifying the client that a reservation of the resource is granted.

Solutions preparing container images and data for container workloads prior to start times of workloads predicted through workload trend analysis. Local storage space on the node is managed based on workload trends, optimizing local storage of image files without requiring frequent reloading and/or deletion of image files, avoiding network intensive I/O operations when pulling images to local storage by workload scheduling systems. Systems perform collection of historical data including image and workload properties; analyze historical data for workload trends, including predicted start times, image files needed, number of nodes and types of nodes. Based on predicted future workload start times, nodes are selected from an ordered list of node requirements and workload properties. Selected nodes' local storage is managed using predicted future start times of workloads, to avoid removing image files having sooner start times, while removing (as needed) images files predictively utilized for workloads further into the future.

Methods, apparatus, systems and articles of manufacture are disclosed for managing compute resources in a computing environment. Disclosed examples are to select an offering workload in a computing environment to lend at least one resource to a needy workload in the computing environment; Disclosed examples are also to cause a host associated with the offering workload to at least one of (i) instantiate a first virtual machine when the host is implemented with a second virtual machine or (ii) instantiate a first container when the host is implemented with a second container. Disclosed examples are further to assign the first virtual machine or the first container to the needy workload.

A system and method for communicating artificial intelligence (AI) tasks between AI resources are provided. The method comprises establishing a connection between a first AI resource and a second AI resource; encapsulating a request to process an AI task in at least one request data frame compliant with a communication protocol, wherein the at least one request data frame is encapsulated at the first AI resource; and transporting the at least one request data frame over a network using a transport protocol to the second AI resource, wherein the transport protocol provisions the transport characteristics of the AI task, and wherein the transport protocol is different than the communication protocol.

A computer based system and method for multilateral computing resource reallocation and asset transaction migration may include: receiving a resource transaction request; determining a policy for the request; identifying, in a resource monitoring database, resources to service the request and choosing resources matching the policy determined for the request; and documenting the choosing of resources in the monitoring database. Embodiments may further include automatically reallocating occupied resources to alternative transactions and/or migrating currently-running tasks to idle resources, for example according to predefined conditions. Embodiments of the invention may allow performing various dynamic, granular computational resource and/or asset reallocation and/or transaction migration procedures which may involve dynamic composition granular individual resources and/or assets (e.g. of multiple types and/or sizes) into functional resources (to be used by, e.g., various workload execution instances) by a resource reallocation hub, which may further include various dedicated modules and/or engines and/or components.

One example method includes determining that local resources at an edge site are inadequate to support performance of a function needed by software running on the edge site, invoking a client agent, in response to invoking the client agent, receiving an execution manifest, determining, by the client agent, where to execute the function, wherein the determining comprises identifying a target execution environment for the function and the determining is based in part on information contained in the execution manifest, and transmitting, by the client agent, the execution manifest to a server agent of the target execution environment, and the execution manifest facilitates execution of the function in the target execution environment.

Techniques for optimizing virtual machine (VM) scheduling on a non-uniform cache access (NUCA) system are provided. In one set of embodiments, a hypervisor of the NUCA system can partition the virtual CPUs of each VM running on the system into logical constructs referred to as last level cache (LLC) groups, where each LLC group is sized to match (or at least not exceed) the LLC domain size of the system. The hypervisor can then place/load balance the virtual CPUs of each VM on the system’s cores in a manner that attempts to keep virtual CPUs which are part of the same LLC group within the same LLC domain, subject to various factors such as compute load, cache contention, and so on.

Described herein are systems, methods, and software to dynamically manage resources across software-defined data centers. In one implementation, a monitoring service obtains flow information associated with physical network interfaces (PNICs) and virtual networking interfaces (VNICs) across a plurality of software-defined data centers (SDDCs). The monitoring service further determines when the flow information associated with the one or more workloads satisfy criteria and, in response to satisfying criteria, generates an update to a configuration associated with at least one SDDC of the plurality of SDDCs based on the flow information.