Streamlining tuple processing by delivering tuple attributes to associated operators
AssigneeInternational Business Machines Corporation
InventorsRyan K. Cradick, Cory J. Kleinheksel
Life of this patent
Estimating the term…
Abstract
Streamlining a stream computing environment by distributing tuple attributes to associated operators in the environment involves receiving a stream of tuples for processing by one or more stream operators. A tuple in the stream of tuples is analyzed to identify one or more data attributes required by the one or more stream operators during processing. Each of the identified attributes is distributed to particular operators that require the identified attribute.
Claims17
A method for processing a stream of tuples comprising:
- receiving a stream of tuples for processing by one or more stream operators;
- analyzing a tuple in the stream of tuples and identifying one or more data attributes of the tuple required by the one or more stream operators during processing, wherein the identifying one or more data attributes of the tuple comprises:
- analyzing code modules of the one or more stream operators to determine data attributes required for computation and data attributes modified by the one or more stream operators;
- and
- comparing data attributes of the tuple to the analysis of the code modules to identify the one or more data attributes;
- distributing at least one of the identified attributes to local storage of particular stream operators of the one or more stream operators that require the identified attributes;
- and
- processing the tuple through the one or more stream operators, wherein the distributed data attributes are processed at the particular stream operators from local storage upon receiving the processed tuple during propagation through the one or more stream operators.
A system for processing a stream of tuples comprising:
- a processor configured to:
- receive a stream of tuples for processing by one or more stream operators;
- analyze a tuple in the stream of tuples and identify one or more data attributes of the tuple required by the one or more stream operators during processing, wherein the identifying one or more data attributes of the tuple comprises:
- analyzing code modules of the one or more stream operators to determine data attributes required for computation and data attributes modified by the one or more stream operators;
- and
- compare data attributes of the tuple to the analysis of the code modules to identify the one or more data attributes;
- distribute at least one of the identified attributes to local storage of particular stream operators of the one or more stream operators that require the identified attributes;
- and
- process the tuple through the one or more stream operators, wherein the distributed data attributes are processed at the particular stream operators from local storage upon receiving the processed tuple during propagation through the one or more stream operators.
A computer program product for processing a stream of tuples, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to:
- receive a stream of tuples for processing by one or more stream operators;
- analyze a tuple in the stream of tuples and identify one or more data attributes of the tuple required by the one or more stream operators during processing, wherein the identifying one or more data attributes of the tuple comprises:
- analyzing code modules of the one or more stream operators to determine data attributes required for computation and data attributes modified by the one or more stream operators;
- and
- comparing data attributes of the tuple to the analysis of the code modules to identify the one or more data attributes;
- distribute at least one of the identified attributes to local storage of particular stream operators of the one or more stream operators that require the identified attributes;
- and
- process the tuple through the one or more stream operators, wherein the distributed data attributes are processed at the particular stream operators from local storage upon receiving the processed tuple during propagation through the one or more stream operators.
Dependencies are inferred from the claim text.