Abstract: A method for adjusting a data parallel region of a stream processing application includes measuring congestion of each parallel channel of the data parallel region, measuring a total throughput of all the parallel channels, and adjusting the number of parallel channels based on the current measured congestion and throughput.

Abstract: A method for adjusting a data parallel region of a stream processing application includes measuring congestion of each parallel channel of the data parallel region, measuring a total throughput of all the parallel channels, and adjusting the number of parallel channels based on the current measured congestion and throughput.

Abstract: Aspects of the present invention provide a solution for recognizing a pattern in a set of data, such as data streaming over a data communication system. In an embodiment, a set of data events is retrieved in the data stream. The retrieved objects each have a plurality of characteristics that can be matched to a predetermined desired characteristic, such as a key value. The retrieved data events can be evaluated with respect to a pattern, with a characteristic of data events being evaluated with respect to an aggregate value related to the pattern. This aggregate value can be updated incrementally based on the data in the characteristic. Based on the evaluation, a determination as to whether the set of data events received subsequent to the first object satisfies the pattern.

Abstract: A method for adjusting a data parallel region of a stream processing application includes measuring congestion of each parallel channel of the data parallel region, measuring a total throughput of all the parallel channels, and adjusting the number of parallel channels based on the current measured congestion and throughput.

Abstract: A method for adjusting a data parallel region of a stream processing application includes measuring congestion of each parallel channel of the data parallel region, measuring a total throughput of all the parallel channels, and adjusting the number of parallel channels based on the current measured congestion and throughput.

Abstract: An embodiment of the invention provides a method for exploiting stateless and stateful data parallelism in a streaming application, wherein a compiler determines whether an operator of the streaming application is safe to parallelize based on a definition of the operator and an instance of the definition. The operator is not safe to parallelize when the operator has selectivity greater than 1, wherein the selectivity is the number of output tuples generated for each input tuple. Parallel regions are formed within the streaming application with the compiler when the operator is safe to parallelize. Synchronization strategies for the parallel regions are determined with the compiler, wherein the synchronization strategies are determined based on the definition of the operator and the instance of the definition. The synchronization strategies of the parallel regions are enforced with a runtime system.

Abstract: An embodiment of the invention provides a method for exploiting stateless and stateful data parallelism in a streaming application, wherein a compiler determines whether an operator of the streaming application is safe to parallelize based on a definition of the operator and an instance of the definition. The operator is not safe to parallelize when the operator has selectivity greater than 1, wherein the selectivity is the number of output tuples generated for each input tuple. Parallel regions are formed within the streaming application with the compiler when the operator is safe to parallelize. Synchronization strategies for the parallel regions are determined with the compiler, wherein the synchronization strategies are determined based on the definition of the operator and the instance of the definition. The synchronization strategies of the parallel regions are enforced with a runtime system.

Abstract: An application provisioning model is automatically created. The model is created from a high-level application and specifies dependencies of the application. It is used to provision the application on one or more nodes or other actions.

Abstract: Data sharing is facilitated in stream processing environments, including distributed stream processing environments. A processor of the stream processing environment obtains at least one of usage information for shared data of the stream processing environment, one or more pre-declared characteristics of the shared data, or performance information relating to the stream processing environment. Based on at least one of the usage information, the one or more pre-declared characteristics or the performance information, code is generated for managing the shared data.

Abstract: A virtual execution environment (VEE) for a streaming Intermediate Language (IL), wherein the streaming IL represents a streaming program, communicates streaming data in queues, stores data-at-rest in variables, and determines data by functions, where inputs are read from the queues and the variables, and outputs are written to the queues and the variables.

Abstract: State sharing is facilitated in stream processing environments, including distributed stream processing environments. A customized shared state implementation representing the state to be shared is automatically created based on at least one of user preferences, hints of usage, and system performance.

Abstract: A method of transforming relational queries of a database into on a data processing system includes receiving a series of relational queries, transforming first parts of the queries into a continuous query embodied as a streaming application, sending parameters in second parts of the queries in the series to the streaming application as a data stream, and executing the continuous query based on the received data stream to generate query results for the series of relational queries. Each query in the series includes a first part and a second part. The first parts are a pattern common to all the queries in the series and the second parts each have one or more parameters that are not common to all of the queries in the series.

Abstract: An Open Database Connectivity (ODBC) proxy infrastructure to transparently route incoming queries to one or more selected query engines. The ODBC proxy receives a query from an application, and determines based on the characteristics of the query and the capabilities of the query engines which one or more query engines are to perform the query. The proxy then routes the query to the one or more query engines, which perform the query. The results are then returned to the proxy, which provides the results to the application.

Abstract: State sharing is facilitated in stream processing environments, including distributed stream processing environments. A customized shared state implementation representing the state to be shared is automatically created based on at least one of user preferences, hints of usage, and system performance.

Abstract: An Open Database Connectivity (ODBC) proxy infrastructure to transparently route incoming queries to one or more selected query engines. The ODBC proxy receives a query from an application, and determines based on the characteristics of the query and the capabilities of the query engines which one or more query engines are to perform the query. The proxy then routes the query to the one or more query engines, which perform the query. The results are then returned to the proxy, which provides the results to the application.

Abstract: A method of transforming relational queries of a database into on a data processing system includes receiving a series of relational queries, transforming first parts of the queries into a continuous query embodied as a streaming application, sending parameters in second parts of the queries in the series to the streaming application as a data stream, and executing the continuous query based on the received data stream to generate query results for the series of relational queries. Each query in the series includes a first part and a second part. The first parts are a pattern common to all the queries in the series and the second parts each have one or more parameters that are not common to all of the queries in the series.

Abstract: State sharing is facilitated in stream processing environments, including distributed stream processing environments. A customized shared state implementation representing the state to be shared is automatically created based on at least one of user preferences, hints of usage, and system performance.

Abstract: Processing streaming data in a data processing system is facilitated by: declaring and defining, by a processor, a request-response interface as part of a stream processing operator defined using a stream processing language; processing a stream of data using the stream processing operator with the request-response interface defined as a part thereof; and communicating with the stream processing operator through the request-response interface via a communication path separate from the stream of data, the communicating accessing or controlling a state of the stream processing operator while the stream processing operator is processing the stream of data.

Abstract: A virtual execution environment (VEE) for a streaming Intermediate Language (IL), wherein the streaming IL represents a streaming program, communicates streaming data in queues, stores data-at-rest in variables, and determines data by functions, where inputs are read from the queues and the variables, and outputs are written to the queues and the variables.

Abstract: An application provisioning model is automatically created. The model is created from a high-level application and specifies dependencies of the application. It is used to provision the application on one or more nodes or other actions.