SUPPLEMENTING A TUPLE WITH A DERIVED VALUE

摘要

A stream of tuples are processed by a plurality of processing elements, each including one or more stream operators, operating on computer processors. A method determines whether a first stream operator includes a first operation that is deterministic and whether a second stream operator includes a second operation that is equivalent to the first operation. In response to determining that the first operation is deterministic and the second operation is equivalent to the first operation, the first stream operator is configured to include in a tuple output from the first stream operator a result of performing the first operation using a first input. The second stream operator, which uses a second input when performing the second operation, determines whether the second input is the same as the first input, and uses the result in lieu of performing the second operation when the first and second inputs are the same.

主权项

1. A system for processing a stream of tuples, comprising:
a plurality of processing elements to receive a stream of tuples, the processing elements operating on one or more computer processors, each processing element having one or more stream operators, the stream operators being arranged in an operator graph; and a component, operating on a computer processor, to: determine whether a first stream operator includes a first operation that is deterministic; determine whether a second stream operator includes a second operation that is equivalent to the first operation; and, in response to determining that the first operation is deterministic and the second operation is equivalent to the first operation, configure by the component the first stream operator to include in a tuple output from the first stream operator a result of performing the first operation using a first input, and configure the second stream operator to:
determine whether a second input is the same as the first input, wherein the second stream operator is configured to use the second input when performing the second operation, anduse the result in lieu of performing the second operation when the second input is the same as the first input.