Techniques for adjusting a compute capacity of a cloud computing system. In an example, a compute scaling application accesses, from a cloud computing system, a compute capacity indicating a number of allocated compute instances of a cloud computing system and usage metrics indicating pending task requests in a queue of the cloud computing system. The compute scaling application determines, for the cloud computing system, a compute scaling adjustment by applying a machine learning model to the compute capability of the cloud computing system and the usage metrics. The compute scaling adjustment indicates an adjustment to a number of compute instances of the cloud computing system. The compute scaling application provides the compute scaling adjustment to the cloud computing system. The cloud computing system adjusts a number of allocated compute instances.

Technologies for allocating resources of managed nodes to workloads to balance multiple resource allocation objectives include an orchestrator server to receive resource allocation objective data indicative of multiple resource allocation objectives to be satisfied. The orchestrator server is additionally to determine an initial assignment of a set of workloads among the managed nodes and receive telemetry data from the managed nodes. The orchestrator server is further to determine, as a function of the telemetry data and the resource allocation objective data, an adjustment to the assignment of the workloads to increase an achievement of at least one of the resource allocation objectives without decreasing an achievement of another of the resource allocation objectives, and apply the adjustments to the assignments of the workloads among the managed nodes as the workloads are performed. Other embodiments are also described and claimed.

An example method is provided for recommending VM configurations, including one or more servers upon which one or more VMs can run. A user wishing to run these VMs can request a recommendation for an appropriate server or set of servers. The user can indicate a category corresponding to the type of workload that pertains to the VMs. The system can receive the request and identify a pool of servers available to the user. Using industry specifications and benchmarks, the system can classify the available servers into multiple categories. Within those categories, similar servers can be clustered and then ranked based on their levels of optimization. The sorted results can be displayed to the user, who can select a particular server (or group of servers) and customize the deployment as needed. This process allows a user to identify and select an optimized setup quickly and accurately.

A bandwidth control method, apparatus, and a device, in the field of computer technologies includes determining an upper bandwidth limit of the device when providing a service for registered clients, resetting an upper bandwidth limit of each client based on a working status of each client and the upper bandwidth limit of the device, and reallocating a bandwidth to each client based on the upper bandwidth limit of each client.

A bandwidth control method, apparatus, and a device, in the field of computer technologies includes determining an upper bandwidth limit of the device when providing a service for registered clients, resetting an upper bandwidth limit of each client based on a working status of each client and the upper bandwidth limit of the device, and reallocating a bandwidth to each client based on the upper bandwidth limit of each client.

Some embodiments of the invention provide a method for evaluating multiple candidate resource elements associated with different resource element types for deploying one tenant deployable element in a single public cloud. The method deploys a set of one or more agents in the public cloud to collect metrics evaluating performance of each of the multiple candidate resource elements. The method communicates with the set of deployed agents to collect metrics to quantify performance of each candidate resource element. The method aggregates the collected metrics in order to generate a report that quantifies performance of each type of candidate resource element for deploying the tenant deployable element in the single public cloud.

Some embodiments of the invention provide a method for evaluating multiple candidate resource elements associated with different resource element types for deploying one tenant deployable element in a single public cloud. The method deploys a set of one or more agents in the public cloud to collect metrics evaluating performance of each of the multiple candidate resource elements. The method communicates with the set of deployed agents to collect metrics to quantify performance of each candidate resource element. The method aggregates the collected metrics in order to generate a report that quantifies performance of each type of candidate resource element for deploying the tenant deployable element in the single public cloud.

Described embodiments provide for routing remote application data. A device can receive a request to access an application. The application can be provided by data centers and accessible via service providers. The device can select a data center from the plurality of data centers and a service provider based at least on a metric indicative of a connection between the data center and the service provider. The device can query a database including one or more connection metrics using the application identified in the request and a location of a router transmitting the request. The device can determine the location of the router based on an internet protocol (IP) address of a client communicably coupled to the router. The device can transmit a response to the request identifying the selected data center and the selected service provider.

Dynamic overflow processing is provided in a multi-user computing environment, which includes receiving, from a user, a request for a new user session at a port of a process of the multi-user computing environment, where the process supports multiple users via the port, and determining that accommodating the new user session will result in resource usage of the process exceeding a predetermined capacity threshold for the process. Based on determining that capacity threshold will be exceeded, the process redirects the request for the new user session to an overflow process started by the process, where the overflow process is an additional instance of the process running within the multi-user computing environment. The process receives a response from the overflow process to the request for the new user session, and forwards the received response to the request for the new user session to the user.

Described herein are systems and methods for end user connection load balancing amongst multiple on-premise connector proxies deployed across geographic locations and reducing connection setup latency without using a shared or distributed database. The system can load balance connections deterministically amongst the on-premise connector proxies using load statistics. The system utilizes an intelligent DNS service that can use network experience data, service availability, and application metrics to provide sophisticated traffic management via DNS or API-based decisions. The system can include a domain name system (DNS) resolver configured to receive metrics for a first connector and a second connector of a data center of an entity, receive a DNS request including an entity identifier and a data center identifier; and transmit a response to the DNS request identifying a server selected based on the metrics identified using the entity identifier and the data center identifier.