StreamClient

This is a Maven project as part of the stream join processing simulation experiments. StreamClient simulates remote stream sources as a command line application.

Requirements

Java
Stream source (dataset) files containing key and value atributes, key will be used as the join attribute

How to use

Build the project (e.g build a StreamClient.jar)
Run the StreamClient.jar within a terminal, e.g. by issuing this command:
```
java -jar StreamClient.jar -h remotehost -p 9999 -d ~/streamSources
```
The command above will read all files inside the ~/streamSources directory path, treating them as the stream sources. Afterwards, as the data being read, the data will be streamed to the remotehost on the port 9999. Here, remotehost is assumed to run the StreamServer command-line application.

Command-line arguments

Below are the complete list of arguments accepted by this application:

h, specify the host target, if not specified localhost is used as the default value
p, specify the host target’s port number, if not specified 4444 is used as the default value
d, specify the source directory for the files which are going to be used as the stream sources, this argument is mandatory. It will read all files inside the directory path specified and stream their contents (assuming key and value attributes) to the remote target. Each of the streams will be identified by its origin filename, e.g. the directory contains rstream.txt and sstream.txt, then the remote target will identify the streams (stream name) as rstream.txt and sstream.txt too.
R{stream_name}={rate_value}, e.g. Rrstream.txt=1000, specify the rate of the data being sent to the remote target (for each of the streams). Rrstream.txt=1000 means that the data in rstream.txt will be sent at the rate of 1000 data/second. Leaving this option blank will assume the stream to use the maximum speed the system could support.