VERFAHREN UND VORRICHTUNG ZUR ASYNCHRONEN BERICHTENÜBERTRAGUNG

PREDICTION-BASED PROVISIONING PLANNING FOR CLOUD ENVIRONMENTS

Various embodiments predict performance of a system including a plurality of server tiers. In one embodiment, a first set of performance information is collected for a base allocation of computing resources across multiple server tiers in the plurality of sever tiers for a set of workloads. A set of experimental allocations of the computing resources is generated on a tier-by-tier basis. Each of the set of experimental allocations varies the computing resources allocated by the base allocation for a single server tier of the multiple server tiers. A second set of performance information associated with the single server tier for each of the set of experimental allocations is collected for a plurality of workloads. At least one performance characteristic of at least one candidate allocation of computing resources across the multiple server tiers is predicted for a given workload based on the first and second sets of performance information. 1. A method , with an information processing system , for predicting performance of a system comprising a plurality of server tiers , the method comprising:collecting, for a base allocation of computing resources across multiple server tiers in the plurality of sever tiers, a first set of performance information associated with the multiple server tiers for a set of workloads;generating, on a tier-by-tier basis, a set of experimental allocations of the computing resources, wherein each of the set of experimental allocations varies the computing resources allocated by the base allocation for a single server tier of the multiple server tiers;collecting, for each of the set of experimental allocations, a second set of performance information associated with the single server tier for a plurality of workloads; andpredicting at least one performance characteristic of one or more candidate allocations of computing resources across the multiple server tiers for a given workload based on the first set of performance information and the ...

VERFAHREN UND VORRICHTUNG ZUR ASYNCHRONEN BERICHTÜBERTRAGUNG

供异步消息接发系统使用的方法和设备

... 一种供异步消息接发系统之用的基于计算机的方法，包括下述步骤：(i)为要在系统中发送的至少一个消息确定至少一种消息传送结果条件，并使所述至少一种消息传送结果条件和要在系统中发送的消息相关联；(ii)在系统中发送具有相关消息传送结果条件的消息；和(iii)监视发送消息的传送，根据相关联消息传送结果条件，评估消息的传送结果，从而可把相对于最终接受者的消息传送结果通知消息的发送者。 ...

Methods and apparatus for use in an asynchronous messaging system

Managing service specifications and the discovery of associated services

Various embodiments manage service labels for association with service specifications in a service-oriented environment. In one embodiment at least one candidate service label is received from a user associated with a service domain of a distributed federated service-oriented environment. The candidate service label includes at least one of a service type and a set of terms characterizing a service. The candidate service label is sent to a community of users associated with one or more service domains in the distributed federated service-oriented environment. One of a vote for accepting the candidate service label and a vote for rejecting the candidate service label is received from each of a plurality of users in the community of users. The candidate service label is stored as an accepted service label in the service domain based on a number of votes for accepting the candidate service label being above a given threshold.

Middleware for externally applied partitioning of applications

A method for routing an application request to servers hosting the application for improved performance and scalability. Routing of the request is accomplished by allocating each partition of an externally defined set of application associated partitions to at least one of the servers hosting the application; by classifying the application request in consideration of its contents according to external criteria; and by assigning the classified application request to one of the partitions; and finally by routing the classified application request to one of said servers hosting the partition.

PREDICTION-BASED PROVISIONING PLANNING FOR CLOUD ENVIRONMENTS

Various embodiments predict performance of a system including a plurality of server tiers. In one embodiment, a first set of performance information is collected for a base allocation of computing resources across multiple server tiers in the plurality of sever tiers for a set of workloads. A set of experimental allocations of the computing resources is generated on a tier-by-tier basis. Each of the set of experimental allocations varies the computing resources allocated by the base allocation for a single server tier of the multiple server tiers. A second set of performance information associated with the single server tier for each of the set of experimental allocations is collected for a plurality of workloads. At least one performance characteristic of at least one candidate allocation of computing resources across the multiple server tiers is predicted for a given workload based on the first and second sets of performance information. 1. A computer program product for predicting performance of a system comprising a plurality of server tiers , the computer program product comprising:a storage medium readable by a processing circuit and storing instructions for execution by the processing circuit for performing a method comprising:collecting, for a base allocation of computing resources across multiple server tiers in the plurality of sever tiers, a first set of performance information associated with the multiple server tiers for a set of workloads;generating, on a tier-by-tier basis, a set of experimental allocations of the computing resources, wherein each of the set of experimental allocations varies the computing resources allocated by the base allocation for a single server tier of the multiple server tiers;collecting, for each of the set of experimental allocations, a second set of performance information associated with the single server tier for a plurality of workloads; andpredicting at least one performance characteristic of one or more candidate ...

Prediction-based provisioning planning for cloud environments

Various embodiments predict performance of a system including a plurality of server tiers. In one embodiment, a first set of performance information is collected for a base allocation of computing resources across multiple server tiers in the plurality of sever tiers for a set of workloads. A set of experimental allocations of the computing resources is generated on a tier-by-tier basis. Each of the set of experimental allocations varies the computing resources allocated by the base allocation for a single server tier of the multiple server tiers. A second set of performance information associated with the single server tier for each of the set of experimental allocations is collected for a plurality of workloads. At least one performance characteristic of at least one candidate allocation of computing resources across the multiple server tiers is predicted for a given workload based on the first and second sets of performance information.

MANAGING SERVICE SPECIFICATIONS AND THE DISCOVERY OF ASSOCIATED SERVICES

Various embodiments manage service labels for association with service specifications in a service-oriented environment. In one embodiment at least one candidate service label is received from a user associated with a service domain of a distributed federated service-oriented environment. The candidate service label includes at least one of a service type and a set of terms characterizing a service. The candidate service label is sent to a community of users associated with one or more service domains in the distributed federated service-oriented environment. One of a vote for accepting the candidate service label and a vote for rejecting the candidate service label is received from each of a plurality of users in the community of users. The candidate service label is stored as an accepted service label in the service domain based on a number of votes for accepting the candidate service label being above a given threshold.

Prediction-based provisioning planning for cloud environments

Various embodiments predict performance of a system including a plurality of server tiers. In one embodiment, a first set of performance information is collected for a base allocation of computing resources across multiple server tiers in the plurality of sever tiers for a set of workloads. A set of experimental allocations of the computing resources is generated on a tier-by-tier basis. Each of the set of experimental allocations varies the computing resources allocated by the base allocation for a single server tier of the multiple server tiers. A second set of performance information associated with the single server tier for each of the set of experimental allocations is collected for a plurality of workloads. At least one performance characteristic of at least one candidate allocation of computing resources across the multiple server tiers is predicted for a given workload based on the first and second sets of performance information.

MANAGING SERVICE SPECIFICATIONS AND THE DISCOVERY OF ASSOCIATED SERVICES

Various embodiments manage service labels for association with service specifications in a service-oriented environment. In one embodiment at least one candidate service label is received from a user associated with a service domain of a distributed federated service-oriented environment. The candidate service label includes at least one of a service type and a set of terms characterizing a service. The candidate service label is sent to a community of users associated with one or more service domains in the distributed federated service-oriented environment. One of a vote for accepting the candidate service label and a vote for rejecting the candidate service label is received from each of a plurality of users in the community of users. The candidate service label is stored as an accepted service label in the service domain based on a number of votes for accepting the candidate service label being above a given threshold.

System and method for grouping diverse operations

A method is provided for managing a group of two or more operations within a software system. The method includes initiating a group context including a sub-context for each operation in the group context, wherein each operation is supported by one of a transactional resource and a messaging component, performing the operations within the sub-contexts, each operation resulting in an outcome, coupling the outcomes within the group context, determining an overall outcome of the group context, and taking at least one action dependent on the overall outcome. The method further includes the step of terminating the group context upon taking one or more actions.

PREDICTION-BASED PROVISIONING PLANNING FOR CLOUD ENVIRONMENTS

Various embodiments predict performance of a system including a plurality of server tiers. In one embodiment, a first set of performance information is collected for a base allocation of computing resources across multiple server tiers in the plurality of sever tiers for a set of workloads. A set of experimental allocations of the computing resources is generated on a tier-by-tier basis. Each of the set of experimental allocations varies the computing resources allocated by the base allocation for a single server tier of the multiple server tiers. A second set of performance information associated with the single server tier for each of the set of experimental allocations is collected for a plurality of workloads. At least one performance characteristic of at least one candidate allocation of computing resources across the multiple server tiers is predicted for a given workload based on the first and second sets of performance information.

Prediction-based provisioning planning for cloud environments

Various embodiments predict performance of a system including a plurality of server tiers. In one embodiment, a first set of performance information is collected for a base allocation of computing resources across multiple server tiers in the plurality of sever tiers for a set of workloads. A set of experimental allocations of the computing resources is generated on a tier-by-tier basis. Each of the set of experimental allocations varies the computing resources allocated by the base allocation for a single server tier of the multiple server tiers. A second set of performance information associated with the single server tier for each of the set of experimental allocations is collected for a plurality of workloads. At least one performance characteristic of at least one candidate allocation of computing resources across the multiple server tiers is predicted for a given workload based on the first and second sets of performance information.

ON-BOARDING SERVICES TO A CLOUD ENVIRONMENT

A system in a cloud services environment comprises one or more service offerings, one or more service instances and one or more service support utilities. Each of the one or more service offerings is described by at least one service descriptor. Each of the one or more service instances is obtained from at least one of the one or more service offerings. Each of the one or more service support utilities is customized by at least one service descriptor. A service comprises at least one component and a service descriptor comprises one or more models. 1. A system in a cloud services environment , comprising:one or more service offerings, each of the one or more service offerings being described by at least one service descriptor;one or more service instances, each of the one or more service instances being obtained from at least one of the one or more service offerings; andone or more service support utilities, each of the one or more service support utilities being customized by at least one service descriptor;wherein a service comprises at least one component and a service descriptor comprises one or more models.2. The system of claim 1 , wherein each of the one or more models of the service descriptor comprises at least one of:an information model that provides identifying information for at least one described service;a structural model that describes one or more components of at least one described service;an operational model that describes a procedure for obtaining at least one service instance of at least one described service;a data model that conveys data items between at least one service offering and at least one service instance of at least one described service;an interaction model that describes a front end user experience at one or more stages in a lifecycle of at least one service instance;a service-level agreement model that allows specification of at least one service quality guarantee agreed to by at least one service offering;a licensing model;a ...

OFFERING APPLICATION PROGRAM INTERFACES (APIS) FOR SALE IN CLOUD MARKETPLACE

To provide a client with access to an API available from a provider, a request is received from a marketplace to subscribe to the API, responsive to the client selecting the API from a catalog of the marketplace. A subscriber identifier, which is uniquely associated with the request by the client to subscribe to the selected API, is generated. An authorization record is created within a proxy; it indicates that a request by the client to invoke the API received by the proxy should be forwarded to the provider from which the API is available if the request to invoke the API includes the subscriber identifier. The generated subscriber identifier and an address of the proxy are provided to the marketplace. Thus, the client is configured to invoke the API by sending a request including the subscriber identifier to the address of the proxy. 1. A method for providing a client with access to an application program interface (API) available from a provider , the method comprising:receiving from a marketplace a request to subscribe to the API, wherein the subscription request is responsive to the client selecting the API from a catalog of the marketplace;generating a subscriber identifier uniquely associated with the request by the client to subscribe to the selected API;creating an authorization record within a proxy, the authorization record indicating that a request by the client to invoke the API received by the proxy should be forwarded to the provider from which the API is available if the request to invoke the API comprises the subscriber identifier;providing the generated subscriber identifier and an address of the proxy to the marketplace, whereby the client is configured to invoke the API by sending a request comprising the subscriber identifier to the address of the proxy.2. The method of claim 1 , wherein creating the authorization record further comprises generating an API key corresponding to the subscriber identifier claim 1 , storing the subscriber identifier ...

AUTOMATIC ABSTRACTION OF FLOW OF CONTROL IN A DISTRIBUTED VIRTUALIZATION PLATFORM

A distributed component software system that includes an analysis server that: (i) receives a set of messages, where each message is sent between components of a distributed component software system (DCSS), and where each message includes an identification of a respective source component, an identification of a respective target component and respective message content; (ii) for each message of the set of received messages, resolves the respective source component; (iii) for each message of the set of received messages, resolves the respective source component; (iv) determines a sequence of the plurality of messages; and (v) generates a human-understandable abstract corresponding to the set of received messages based upon the respective message sources, the respective message targets, the respective message content and the sequence of the set of received messages. 1. A method comprising:receiving, by an analysis server, a plurality of messages, where each message of the plurality of messages is sent between components of a distributed component software system (DCSS), and where each message includes an identification of a respective source component, an identification of a respective target component and respective message content;for each message of the plurality of messages, resolving, by the analysis server, the respective source component;determining, by the analysis server, a sequence of the plurality of messages; andgenerating, by the analysis server, a human-understandable abstract corresponding to the plurality of messages based upon the respective message sources, the respective message targets, the respective message content and the sequence of the plurality of messages.2. The method of further comprising:presenting the abstract to a human user.3. The method of further comprising:saving the abstract in machine readable form.4. The method of wherein each message of the plurality of messages is in a respective message communication protocol of a plurality ...

Release and management of composite applications on paas

A solution descriptor comprises a set of component workload units, a workload unit describing a deployable application component with application binary, configuration parameters and dependency declarations. An environment descriptor specifies a set of target platforms and plugins in an execution environment. A deployer interprets the solution descriptor and the environment descriptor, and generates a list of tuples comprising compatible workload-plugin-platform combinations. The deployer determines an execution order for the list of tuples, and invokes the plugins in the list of tuples in the execution order, wherein each of the plugins executes a corresponding compatible workload on a corresponding compatible target platform specified in the list of tuples.

Non-Intrusive Cloud Services Billing

An embodiment of the invention provides a method for non-intrusive cloud services billing, including storing service log files in a data store, wherein each of the service log files include a lifecycle event of a resource in a service produced by an instance of the service. A log transformer connected to the data store converts the service log files into a standardized format. A metering descriptor describing billable entities for each offering of the service is obtained, wherein each offering of the service has a different metering descriptor, and wherein the billable entities include multiple categories of billing. A billing record generator connected to the log transformer is programmed, wherein the billing record generator converts standardized log records into metering records. The billing record generator generates a metering record with one or more of the service log files in the standardized format and the metering descriptor. 1. A method comprising:storing service log files in a data store, each of the service log files including a lifecycle event of a resource in a service produced by an instance of the service;converting the service log files into a standardized format;obtaining a metering descriptor describing billable entities for each offering of the service, wherein the billable entities include multiple categories of billing; andgenerating a metering record with the billing record generator, wherein the metering record is generated with at least one of the service log files in the standardized format and the metering descriptor.2. The method according to claim 1 , wherein the metering record includes a billable entity claim 1 , a event claim 1 , and event data.3. The method according to claim 2 , further comprising assigning a rating to the metering record based on the billable entity.4. The method according to claim 3 , further comprising generating an invoice for the metering record based on the rating.5. The method according to claim 1 , further ...

RELEASE AND MANAGEMENT OF COMPOSITE APPLICATIONS ON PAAS

A solution descriptor comprises a set of component workload units, a workload unit describing a deployable application component with application binary, configuration parameters and dependency declarations. An environment descriptor specifies a set of target platforms and plugins in an execution environment. A deployer interprets the solution descriptor and the environment descriptor, and generates a list of tuples comprising compatible workload-plugin-platform combinations. The deployer determines an execution order for the list of tuples, and invokes the plugins in the list of tuples in the execution order, wherein each of the plugins executes a corresponding compatible workload on a corresponding compatible target platform specified in the list of tuples. 1. A method of specifying , deploying and managing composite applications , comprising:receiving a solution descriptor comprising a set of component workload units, a component workload unit describing a deployable application component with application binary, configuration parameters and dependency declarations;receiving an environment descriptor specifying a set of target platforms and plugins available in an execution environment;generating, by a processor, based on the solution descriptor and the environment descriptor, a list of tuples comprising compatible workload-plugin-platform combinations;determining, by the processor, an execution order for the list of tuples; andinvoking, by the processor, the plugins in the list of tuples in the execution order, each of the plugins executing a corresponding compatible component workload unit on a corresponding compatible target platform specified in the list of tuples,wherein the set of component workload units comprises an instance of a component application type in a set of component application types, and the set of component application types is extensible.2. The method of claim 1 , wherein the set of component application types comprises one or more of ...

RELEASE AND MANAGEMENT OF COMPOSITE APPLICATIONS ON PAAS

A solution descriptor comprises a set of component workload units, a workload unit describing a deployable application component with application binary, configuration parameters and dependency declarations. An environment descriptor specifies a set of target platforms and plugins in an execution environment. A deployer interprets the solution descriptor and the environment descriptor, and generates a list of tuples comprising compatible workload-plugin-platform combinations. The deployer determines an execution order for the list of tuples, and invokes the plugins in the list of tuples in the execution order, wherein each of the plugins executes a corresponding compatible workload on a corresponding compatible target platform specified in the list of tuples. 1. A system for the description , deployment and management of composite applications , the system comprising:a processor;a memory device coupled to the processor;a solution descriptor, stored on the memory device, comprising a set of component workload units, a component workload unit describing a deployable application component with application binary, configuration parameters and dependency declarations;an environment descriptor, stored on the memory device, specifying a set of target platforms and plugins in an execution environment; anda deployer running on the processor, and operable to generate, based on the solution descriptor and the environment descriptor, a list of tuples comprising compatible workload-plugin-platform combinations, the deployer further operable to determine an execution order for the list of tuples, the deployer further operable to invoke the plugins in the list of tuples in the execution order, each of the plugins executing a corresponding compatible component workload unit on a corresponding compatible target platform specified in the list of tuples,wherein the set of component workload units comprises an instance of a component application type in a set of component application ...

System and method for caching sets of objects

A system and method for managing cacheable sets of objects having explicit lifetime specifications, wherein a time-based cache manager maintains and updates one or more sets of objects stored in the cache. A cached set of objects comprises objects having start times and/or end times representing, respectively, times at which such objects become valid and expire. An update time is determined for a given cached set of objects based, in part, on the start times and end times of objects comprising the cached set of objects. When a request for the retrieval of an object from the given cached set of objects (or the entire cached set) is received, a determination is made if the cached set of objects is valid at the time of the request based on the update time. If the cached set of objects is not valid because, e.g., the update time has elapsed, the cached set of objects is updated (and any other cached sets, if necessary) by deleting and/or adding objects to the cached set of objects having start times and end times that meet predefined time criteria for inclusion in the cached set of objects.

System and method for employing externalized, dynamically configurable, cacheable trigger points

There is provided a method for employing a trigger point in a logic flow. The trigger point has at least one function associated therewith. A default strategy set is associated with the at least one function of the trigger point. The default strategy set has a capability of being replaced, in whole or in part, by a non-default strategy set. The strategy sets are for implementing the at least one function of the trigger point. The trigger point is dynamically configured, including establishing at least one context for the trigger point that respectively specifies a location of the strategy sets. The trigger point is executed when encountered during an execution of the logic flow, including selectively executing at least a portion of at least one of the default strategy set and the non-default strategy set based upon a current context from among the at least one context.

Automated deployment of applications with tenant-isolation requirements

An embodiment of the invention provides a method for multi-tenant enabling a service, where the service includes a plurality of service components. Rules are received from a provider of the service, wherein the rules include rules for isolating the service components and rules for sharing the service components. The service components are matched with virtualization environments and/or physical environments based on the rules. The virtualization environments include at least one isolation container and/or at least one shared container. The service components are stored in the virtualization environments and/or the physical environments based on the matching. The service components are mapped to the at least one isolation container of the virtualization environments, the at least one shared container of the virtualization environments, and/or the physical environments based on the storing of the service components.

System and method for caching sets of objects

A system and method for managing cacheable sets of objects having explicit lifetime specifications, wherein a time-based cache manager maintains and updates one or more sets of objects stored in the cache. A cached set of objects comprises objects having start times and/or end times representing, respectively, times at which such objects become valid and expire. An update time is determined for a given cached set of objects based, in part, on the start times and end times of objects comprising the cached set of objects. When a request for the retrieval of an object from the given cached set of objects (or the entire cached set) is received, a determination is made if the cached set of objects is valid at the time of the request based on the update time. If the cached set of objects is not valid because, e.g., the update time has elapsed, the cached set of objects is updated (and any other cached sets, if necessary) by deleting and/or adding objects to the cached set of objects having start times and end times that meet predefined time criteria for inclusion in the cached set of objects.

Method and apparatus for deploying a set of virtual software resource templates to a set of nodes

The illustrative embodiments described herein provide a computer-implemented method, apparatus, and computer program product for deploying a set of virtual software resource templates to a plurality of nodes. The process identifies a set of virtual machine images and metadata to form a set of virtual software resource templates. The process creates a model that indicates a performance metric based on a number of nodes on which a number of virtual machine images is deployed. The process receives a quality of service requirement. The process determines, using the model, a sufficient number of virtual software resource templates in the set of virtual software resource templates and a sufficient number of nodes in the plurality of nodes to fulfill the quality of service requirement. The process deploys the sufficient number of virtual software resource templates on the sufficient number of nodes using the metadata.

Service Description Refinement Based on Actual Service Use

Techniques are disclosed for generation and refinement of service descriptions based on records of invocations of the services, i.e., service description refinement based on actual service use. For example, a method for describing one or more services in a service-oriented environment comprised of one or more clients invoking the one or more services comprises the following steps. An initial description is established for at least one of the one or more services. Information is collected from one or more service invocations including at least one of a client identity, a value of at least one parameter, a return value, and an execution time. The information is analyzed to refine the description.

Methods and apparatus for business rules authoring and operation employing a customizable vocabulary

Methods and apparatus for authoring and executing an individualized language business rule. In one embodiment, a method comprises creating at least one individualized language resource, creating at least one individualized language rule referencing at least one of said individualized language resource, and transforming said at least one individualized language rule into computer executable format.

Methods and apparatus for access control in service-oriented computing environments

Improved access control techniques for use in a service-oriented computing environment are disclosed. For example, one method for authenticating a client in a service-oriented environment, wherein the service-oriented environment includes a plurality of services, includes the following steps. At least one service of the plurality of services is invoked. State information is associated with the at least one service invoked. The state information is used to authenticate a client with at least one service. Further, a method for access control in a service-oriented environment, wherein the service-oriented environment includes a plurality of services, includes the following steps. A rule specification language is provided. At least one rule is specified using the rule specification language. A verification is performed to determine whether or not the client satisfies the at least one rule. The client is granted access to a service when the client satisfies the at least one rule.

On-boarding services to a cloud environment

A system in a cloud services environment comprises one or more service offerings, one or more service instances and one or more service support utilities. Each of the one or more service offerings is described by at least one service descriptor. Each of the one or more service instances is obtained from at least one of the one or more service offerings. Each of the one or more service support utilities is customized by at least one service descriptor. A service comprises at least one component and a service descriptor comprises one or more models.

System and method for managing cachable entities

A system and method for managing cachable entities (i.e., entities stored in a cache and/or entities which may be stored in a cache) in a data processing application. In one aspect, a method for managing cachable entities comprises the steps of analyzing program code to determine if there is at least one statement which affects a desirability of performing at least one cache transaction; and performing the at least one cache transaction if it is desired. In another aspect, a program analysis tool statically analyzes program code to locate points where object state changes occur, where objects are created and where objects are deleted, and then generates regularized dependencies at such points for and employing the dependencies to invalidate dependent cached queries. Query specific keys are generated to insert query results into and retrieve query results from a dependency managed cache. Regularized dependencies are generated at the object query points and are attached to query results inserted into a dependency managed cache, which facilitates consistency of query results from the dependency managed cache.

System and method for externalizable inferencing components

Method and apparatus for applying business rules in an object model driven context

A means is provided for specifying, applying, and managing sets of temporary or permanent additions or modifications to the behavior of object-oriented programs without having to change the code of the program, by using externalized rules. The points at which the externalized rules may be applied is determined by the implementation object model, thus making their specification natural to the program developers who are familiar with the program's implementation object model. The concept of dynamic method-based trigger or control points is disclosed for identifying potential rule attachment points in objects and identifying the rules that are applicable to each dynamic control point.

Methods and apparatus for asynchronous messaging

A computer-based method for use in an asynchronous messaging system comprises the steps of: (i) defining at least one message delivery outcome condition for at least one message to be sent in the system and associating the message delivery outcome condition with the message to be sent in the system; (ii) sending the message with the associated message delivery outcome condition in the system; and (iii) monitoring the delivery of the sent message and evaluating the delivery outcome of the message with respect to the associated message delivery outcome condition such that a sender of the message may be notified as to the outcome of the delivery of the message to final recipients.

Method and data processing system for specifying and applying rules to classification-based decision points in an application system

A method and apparatus for specifying, applying and managing rules used by an application in a data processing system are provided. A set of classification categories are specified, each classification category representing a different purpose of classification. A set of classifications are specified, each classification representing a possible outcome of an act of classifying. A set of control point names are specified, each name being associated with one or more control points in the application. A set of classification rules are specified, each such rule being associated with a classification category and said rule being adapted to analyze the state of the application and classify it by returning one or more classifications. A set of decision rules are specified, each such rule being associated with a classification and a control point name, said rule being adapted to affect the behavior of the application by calculating a value or making a decision. A set of control points is specified and built into the application at those points in the application flow at which variability of behavior controlled by rules is desired, each such control point being associated with a control point name and one or more classification categories. During the running of the application, when a control point as described above is encountered, the control point's associated classification categories are used to select that set of classification rules associated with any of the classification categories and the selected classification rules are then run.

Methods and apparatus for coordinating a plurality of updates to a cache

Techniques for improved cache management are provided. In particular, the invention provides techniques for improving the task of updating a cache by efficiently and intelligently combining and controlling cache update operations. For example, a technique may comprise the following steps. A set of keys is determined such that at least one key in the set corresponds to an object in an unknown state. Requests to keys in the set are suspended while the cache is being updated. That is, the set of keys is set to an offline status, i.e., not accepting any request for a key in the set. Then, the set is placed back to an online status, i.e., accepting requests for a key in the set, after the update to the cache is complete.

Release and management of composite applications on PaaS

A solution descriptor comprises a set of component workload units, a workload unit describing a deployable application component with application binary, configuration parameters and dependency declarations. An environment descriptor specifies a set of target platforms and plugins in an execution environment. A deployer interprets the solution descriptor and the environment descriptor, and generates a list of tuples comprising compatible workload-plugin-platform combinations. The deployer determines an execution order for the list of tuples, and invokes the plugins in the list of tuples in the execution order, wherein each of the plugins executes a corresponding compatible workload on a corresponding compatible target platform specified in the list of tuples.

Release and management of composite applications on PAAS

A solution descriptor comprises a set of component workload units, a workload unit describing a deployable application component with application binary, configuration parameters and dependency declarations. An environment descriptor specifies a set of target platforms and plugins in an execution environment. A deployer interprets the solution descriptor and the environment descriptor, and generates a list of tuples comprising compatible workload-plugin-platform combinations. The deployer determines an execution order for the list of tuples, and invokes the plugins in the list of tuples in the execution order, wherein each of the plugins executes a corresponding compatible workload on a corresponding compatible target platform specified in the list of tuples.

Release and management of composite applications on PaaS

A solution descriptor comprises a set of component workload units, a workload unit describing a deployable application component with application binary, configuration parameters and dependency declarations. An environment descriptor specifies a set of target platforms and plugins in an execution environment. A deployer interprets the solution descriptor and the environment descriptor, and generates a list of tuples comprising compatible workload-plugin-platform combinations. The deployer determines an execution order for the list of tuples, and invokes the plugins in the list of tuples in the execution order, wherein each of the plugins executes a corresponding compatible workload on a corresponding compatible target platform specified in the list of tuples.