Quality-driven processing of out-of-order data streams

A non-transitory, computer-readable medium storing computer-readable instructions executable by a computer, wherein the instructions instruct the computer to:

• identify, by the computer, a configurable quality metric defined by a user or application for executing continuous queries associated with a data stream over a sliding window, the configurable quality metric specifying a user requirement for the quality of the continuous query result;
• calculate, by the computer, a quality threshold metric based on the identified configurable quality metric, the configurable quality metric associated with a size of an adaptation buffer for incoming tuples associated with the continuous query associated with the data stream;
• in response to receiving a signal indicating a query result is generated, identify, by the computer, a sliding window prior to a current time;
• measure, by the computer, a runtime quality metric associated with the incoming tuples for the identified sliding window;
• and
• dynamically modify, by the computer, the adaptation buffer at query runtime based on a difference between the quality threshold metric and the measured runtime quality metric.

A system, comprising:

• a memory;
• at least one hardware processor interoperably coupled with the memory and configured to:;
• identify, by the at least one hardware processor, a configurable quality metric defined by a user or application for executing continuous queries associated with a data stream over a sliding window, the configurable quality metric specifying a user requirement for the quality of the continuous query result;
• calculate, by the at least one hardware processor, a quality threshold metric based on the identified configurable quality metric, the configurable quality metric associated with a size of an adaptation buffer for incoming tuples associated with the continuous query associated with the data stream;
• in response to receiving a signal indicating a query result is generated, identify, by the at least one hardware processor, a sliding window prior to a current time;
• measure, by the at least one hardware processor, a runtime quality metric associated with the incoming tuples for the identified sliding window;
• and
• dynamically modify, by the at least one hardware processor, the adaptation buffer at query runtime based on a difference between the quality threshold metric and the measured runtime quality metric.