Abstract: A configuration is generated for a software that is to be deployed for providing high service availability to satisfy configuration requirements. One or more configuration patterns are identified, each of which specifies a set of attribute values and an actual recovery action for a failed component as a configuration option of the software. The unchangeable attribute values of the software are matched with the configuration patterns to obtain a matching configuration pattern, whose actual recovery action incurs a smallest component failure recovery impact zone. The matching configuration pattern is selected as at least a portion of the configuration of the software. Then the changeable attribute values of the software are set to the corresponding attribute values of the matching configuration pattern to satisfy the configuration requirements.

Abstract: Disjoint node groups are generated for a node cluster to guarantee hardware redundancy for each service group to be configured on the node cluster. Each service group includes service units, and each service unit is a unit of redundancy for services provided and protected by the service group. Hardware dependency is identified between virtual machines in the node cluster and hardware elements to host the virtual machines. A first node group set (NG set) is formed using a first criterion to select the virtual machines into different node groups based on the hardware dependency, such that there is no common hardware element between any two node groups in the first NG set. The first NG set is mapped to a first set of service groups. Different node groups of the first NG set are mapped to different service units in each service group of the first set of service groups.

Abstract: An embedded step is executed to upgrade multiple layers of a system. The embedded step includes multiple nested steps, and each nested step includes a sequence of actions for upgrading one of the layers. A finite state machine defines the state transitions for each nested step. During the execution of the nested steps, an outer step of the nested steps is transitioned from an executing state into a waiting state after the outer step completes a tear-down. During the transitioning of the outer step, the outer step sends a trigger to an inner step of the nested steps to cause the inner step to transition into the executing state. When the inner step is completed, the outer step is transitioned out of the waiting state to execute a build-up phase. Execution of the embedded step is completed when an outermost step of the nested steps is completed.

Abstract: An upgrade campaign is generated for software deployment configuration, based on a source configuration and a target configuration for a software subsystem. A delta between the source configuration and the target configuration is determined. The delta includes a first set, a second set and a third set of the entities to be removed, added and upgraded, respectively, in the software subsystem. Based on dependencies among the entities, a sequence of stages is generated for removing, adding and upgrading the first, second and third sets of the entities, such that the entities are removed, added and upgraded according to a temporal order defined by the sequence of stages.

Abstract: A configuration is generated that enables Availability Management Framework (AMF) to manage virtual machines (VMs) and support hardware redundancy for each of service groups (SGs). Each component service instance (CSI) is configured to represent one or more of the VMs to be managed by AMF. Configuration attributes are set to associate each VM with one CSI, such that the VMs to be hosted on a same physical host are associated with different CSIs of a same service instance. Service units (SUs) of different SGs are configured such that each SU is hosted by any one VM on the same physical host, while the different SUs of a same SG are mapped to the VMs hosted on different physical hosts. The configuration enables the AMF to manage an application formed by the SGs in a virtualized environment in a same way as in a non-virtualized environment without modification to the application.

Abstract: Configuration requirements that specify the provision of services using a system-level description are automatically generated from user requirements. The user requirements are decomposed into one or more levels of decomposed functionalities using an ontology as input. The ontology stores known decompositions of functionalities and relations between the known decompositions. The lowest level of the decomposed functionalities is mapped into a set of components provided by vendors, and additional components on which the set of components depend are identified. Based on the set of components and the additional components, a required number of instances of service workload is calculated to generate the configuration requirements of the system that satisfy the user requirements.

Abstract: Redundancy models are compared to determine or assist in determining an Availability Management Framework (AMF) configuration of a highly available system based on quantified service availability of the system. Each redundancy model defines assignments of service-instances to service-units. An analysis model of the system is constructed to capture recovery behaviors of the system for each redundancy model. Service availability of the system is quantified based on the analysis model under one or more scenarios including failure scenarios and recovery scenarios. Based on a comparison of service availability levels provided by the redundancy models and subject to constraints of the HA system, one of the redundancy models is identified that provides a required level of service availability for the system.

Abstract: An Availability Management Framework (AMF) configuration is generated, in which service units of a service group are allocated to node groups and the node groups are allocated to physical nodes. Each service unit is a unit of redundancy for services provided and protected by the service group. Each node group includes a group of virtual machines. Each node group is mapped to a different host group that includes one or more physical nodes. Collocation and separation rules are generated for the virtual machines to specify which virtual machines are collocated in a same node group and which virtual machines are separated by different node groups. Each service unit of the service group is mapped to a different node group to guarantee hardware separation in the AMF configuration for different service units of the service group and for the virtual machines allocated to the different service units.

Abstract: Techniques for mapping availability management (AM) functions to software installation locations are described. An availability management function (AMF) can look-up a component type and determine software associated with that component. For a selected AMF node, the AMF software entity can then determine a pathname prefix associated with that software. The pathname prefix can then be used for various AM functions, e.g., instantiation of a new component or service unit.

Abstract: Disjoint node groups are generated for a node cluster to guarantee hardware redundancy for each service group to be configured on the node cluster. Each service group includes service units, and each service unit is a unit of redundancy for services provided and protected by the service group. Hardware dependency is identified between virtual machines in the node cluster and hardware elements to host the virtual machines. A first node group set (NG set) is formed using a first criterion to select the virtual machines into different node groups based on the hardware dependency, such that there is no common hardware element between any two node groups in the first NG set. The first NG set is mapped to a first set of service groups. Different node groups of the first NG set are mapped to different service units in each service group of the first set of service groups.

Abstract: A method and system for generating an Availability Management Framework (AMF) configuration based on a model driven approach. The AMF configuration is an instance of an AMF sub-profile that can be used to model resources and services to be protected, and is generated from an instance of the Entity Type Files (ETF) sub-profile and an instance of the Configuration Requirements (CR) sub-profile. The ETF sub-profile can be used to model the resources provided by vendors, and the CR sub-profile can be used to model configuration requirements. Each of the AMF sub-profile, the ETF sub-profile and the CR sub-profile is a specialization of pre-defined Unified Modeling Language (UML) meta-classes. An input that includes an ETF model and a CR model, which are instances of the ETF sub-profile and the CR sub-profile, respectively, is transformed into an AMF model as the AMF configuration.

Abstract: An Availability Management Framework (AMF) configuration is generated, in which service units of a service group are allocated to node groups and the node groups are allocated to physical nodes. Each service unit is a unit of redundancy for services provided and protected by the service group. Each node group includes a group of virtual machines. Each node group is mapped to a different host group that includes one or more physical nodes. Collocation and separation rules are generated for the virtual machines to specify which virtual machines are collocated in a same node group and which virtual machines are separated by different node groups. Each service unit of the service group is mapped to a different node group to guarantee hardware separation in the AMF configuration for different service units of the service group and for the virtual machines allocated to the different service units.

Abstract: An Availability Management Framework (AMF) configuration describes how configuration entities of a highly available system are grouped and includes information on service provision and service protection policies against resource failure. The AMF configuration defines a set of failure types for each component and each node, and specifies a failure rate and a recommended recovery for each failure type. A method for evaluating service availability receives the AMF configuration as input, and analyzes it to obtain an actual recovery that the highly available system is to perform when the given component fails. The method maps the AMF configuration to a stochastic model that captures the dependencies among the components and among the configuration entities at multiple levels of the hierarchy. The method utilizes the model to calculate the service availability of the AMF configuration based on the failure rate, the actual recovery and the dependencies.

Abstract: Presented are methods and apparatus for protecting a plurality of High Availability (HA) Service Instances (SIs) with a plurality of Service Units (SUs) with an Nway redundancy model. Any of the SUs associated with the Nway redundancy model can simultaneously be assigned an active HA state for some of the SIs and a standby HA state for other SIs. However, only one SU can have the active state for any given SI. The Nway redundancy model is a configured prior to runtime operation.

Abstract: Configuration requirements that specify the provision of services using a system-level description are automatically generated from user requirements. The user requirements are decomposed into one or more levels of decomposed functionalities using an ontology as input. The ontology stores known decompositions of functionalities and relations between the known decompositions. The lowest level of the decomposed functionalities is mapped into a set of components provided by vendors, and additional components on which the set of components depend are identified. Based on the set of components and the additional components, a required number of instances of service workload is calculated to generate the configuration requirements of the system that satisfy the user requirements.

Abstract: Among other things, embodiments described herein enable systems, e.g., Availability Management Forum (AMF) systems, having service units to operate with balanced loads both before and after the failure of one of the service units. A configuration can be generated which provides for distributed backup roles and balanced active loads. When a failure of a service unit occurs, the active loads previously handled by that service unit are substantially evenly picked up as active loads by remaining service units.

Abstract: Configuration requirements for an Availability Management Framework (AMF) configuration are generated from high level user requirements that specify a subset of properties of an AMF configuration. The user requirements are first mapped into entity prototypes defined in an extended Entity Types File (ETF) model. A computer system identifies additional entity prototypes and dependency thereof that support functionalities of the mapped entity prototypes under a set of grouping conditions. The computer system calculates the required number of component service instances (CSIs) and service instances (SIs) that satisfy the user requirements. The calculation is based on the subset of the properties specified by the user requirements and measurements associated with the entity prototypes in the extended ETF model. The computer system then forms configuration requirements including the required number of CSIs and SIs for generating the AMF configuration that satisfies the user requirements.

Abstract: Redundancy models are compared to determine or assist in determining an Availability Management Framework (AMF) configuration of a highly available system based on quantified service availability of the system. Each redundancy model defines assignments of service-instances to service-units. An analysis model of the system is constructed to capture recovery behaviors of the system for each redundancy model. Service availability of the system is quantified based on the analysis model under one or more scenarios including failure scenarios and recovery scenarios. Based on a comparison of service availability levels provided by the redundancy models and subject to constraints of the HA system, one of the redundancy models is identified that provides a required level of service availability for the system.

Abstract: A configuration including Service Instances (SIs) and a list of Service Units (SUs) is to be validated. The SIs are to be allocated to the SUs for protection of the service represented by the SIs. A set of heuristics is applied to determine whether, for each of the SI assignments, the SI can be allocated to one of the SUs whose capacities support the required capacities of the SI. The heuristic then walks the list in order, to find a first SU that supports a current SI. If none of the SUs in the list can support the current SI, the heuristic indicates that the configuration is not validated. In response to a result that at least one of the heuristics in the set indicates the SUs can support all of the SIs, a final result is generated indicating that the configuration is valid.

Abstract: Among other things, embodiments described herein enable systems, e.g., Availability Management Forum (AMF) systems, having service units to operate with balanced loads both before and after the failure of one of the service units. A method for balancing standby workload assignments and active workload assignments for a group of service units in a system which employs an N+M redundancy model, wherein N service units are active service units and M service units are standby service units is described. An active workload that the N active service units need to handle is calculated and each of the N active service units in the group is provided with an active workload assignment based on the calculated active workload. Standby workload assignments are distributed among the M standby service units substantially equally.