The core of GeDDSProCon. Geo-distributed data stream processor connector allows user to connect geo-distributed heterogeneous data stream processors (Flink, Spark). As of now it is only possible to use the data stream processors Apache Spark and Apache Flink.
These classes are used for our evaluations (local and in the CLoud). Each DSP implements the reference pipelines (Ref-Pipeline_A, Ref-Pipeline_B) and distributed pipelines (Dis-Pipeline_A, Dis-Pipeline_B).
The message-buffer is a separate process and runs besides the data stream processing pipeline and is used to buffer all data which is sent/received by the pipeline.
Consists of the word-count examples of the supporting data stream processors. Can be used to understand how
geddsprocon can be used.