Systems and methods directing requests to a particular node in a multi-tiered middleware environment are provided. In one embodiment, traffic requests can be dynamically routed to a desired server, and not be load balanced, based on an identification of the client generating the traffic requests.

Active-active standby is maintained for communication sessions using web connections between two active session zones and a standby session zone, for example, in the event of a failure. The first active session zone is located at a first datacenter, the second active session zone is located at a second datacenter, and the standby session zone is located at a third datacenter. In the event of a failure at a first active session zone at the first datacenter, a failover to the second active session zone at the second datacenter is performed. In the event of a failure at both active session zones, a failover to the standby session zone at the third datacenter is performed.

Active-active standby is maintained for communication sessions using session initiation protocol (SIP) processes between two active session zones in a first datacenter and a standby session zone in a second datacenter. In the event of a failure at a first active session zone at the first datacenter, a failover to the second active session zone at the first datacenter is performed such that there are no interruptions in the active sessions. In the event of a failure at both active session zones at the first datacenter, a failover to the second datacenter is performed.

Techniques are described for enabling a cloud-based IT and security operations application to execute playbooks containing custom code in a manner that mitigates types of risk related to the misuse of cloud-based resources and security of user data. Users use a client application to create and modify playbooks and, upon receiving input to save a playbook, the client application determines whether the playbook includes custom code. If the client application determines that the playbook includes custom code, the client application establishes a connection with a proxy application (also referred to as an “automation broker”) running in the user's own on-premises network and sends a representation of the playbook to the proxy application. The client application further sends to the IT and security operations application an identifier of the playbook and an indication that the playbook (or the custom code portions of the playbook) is stored within the user's on-premises network.

A computerized method of managing a computer remote session operation, comprising providing a server for hosting application execution; configuring a number of predefined user accounts with low security permissions on said server, where said user accounts are not tied to any specific real user; Whenever a remote user requests to start a remote session, finding an available user account not currently in use on said computer, allocating it for the remote session and marking it as unavailable for subsequent session requests; Generating a one-time password for said user account; Communicating the assigned user account identifier and temporary password to client component on the user's side, either directly or through an intermediate broker; causing the client component to connect to the server using said user account identifier and temporary password; and, upon termination of the remote session, deleting the assigned user account's data and marking it as available again.

A method and apparatus for differentially optimizing a quality of service (QoS) includes: establishing a system model of a multi-task unloading framework; acquiring a mode for users executing a computation task, executing, according to the mode for users executing the computation task, the system model of the multi-task unloading framework; and optimizing a quality of service (QoS) on the basis of a multi-objective optimization method for a multi-agent deep reinforcement learning. According to the present invention, an unloading policy is calculated on the basis of a multi-user differentiated QoS of a multi-agent deep reinforcement learning, and with the differentiated QoS requirements among different users in a system being considered, a global unloading decision is performed according to a task performance requirement and a network resource state, and differentiated performance optimization is performed on different user requirements, thereby effectively improving a system resource utilization rate and a user service quality.

A method for routing requests to a plurality of server clusters is disclosed. The method comprises establishing a first server cluster responding to requests concerning a first software via a first version of that software and to requests concerning a second software via a first version of that software, and a second server cluster responding to requests concerning the first software via a second version of that software and to requests concerning the second software via a second version of that software. A gateway router initially routes requests concerning the first and second software by default to the first server cluster. Upon receiving a request to change default routing of requests, a configuration of the gateway router is updated. Subsequent requests concerning the first software are routed to the second server cluster while subsequent request concerning the second software remain routed to the first server cluster.

A service request allocation method is provided, performed by a computer device, the method including receiving a service request that needs to be processed; dividing a service device set into a plurality of service device subsets, the plurality of service device subsets being grouped based on an idle resource of each service device in the service device set; determining a target service device subset that matches the service request according to an allocation weight corresponding to each service device subset; selecting a target service device from the target service device subset; and allocating the service request to the target service device, the target service device being configured to perform a service corresponding to the allocated service request.

A network edge computing method includes receiving, by an edge data node, a service request at least processed by network edge computation scheduling; and routing, according to a service port involved in the service request, the service request to a container of the edge data node, to be processed by the container.

A network edge computing method includes receiving, by an edge data node, a service request at least processed by network edge computation scheduling; and routing, according to a service port involved in the service request, the service request to a container of the edge data node, to be processed by the container.