diff options
Diffstat (limited to 'wordcount/resources/topology.yaml')
| -rw-r--r-- | wordcount/resources/topology.yaml | 73 |
1 files changed, 73 insertions, 0 deletions
diff --git a/wordcount/resources/topology.yaml b/wordcount/resources/topology.yaml new file mode 100644 index 0000000..725f3be --- /dev/null +++ b/wordcount/resources/topology.yaml @@ -0,0 +1,73 @@ +# topology definition + +# name to be used when submitting. This is what shows up... +# in the Storm UI/storm command-line tool as the topology name +# when submitted to Storm +name: "wordcount" + +# Topology configuration +config: + # Hint for the number of workers to create + topology.workers: 1 + +# Spout definitions +spouts: + - id: "sentence-spout" + className: "org.apache.storm.flux.wrappers.spouts.FluxShellSpout" + constructorArgs: + # Command line + - ["python", "sentencespout.py"] + # Output field(s) + - ["sentence"] + # parallelism hint + parallelism: 1 + +# Bolt definitions +bolts: + - id: "splitter-bolt" + className: "org.apache.storm.flux.wrappers.bolts.FluxShellBolt" + constructorArgs: + # Command line + - ["python", "splitbolt.py"] + # Output field(s) + - ["word"] + parallelism: 1 + + - id: "counter-bolt" + className: "org.apache.storm.flux.wrappers.bolts.FluxShellBolt" + constructorArgs: + # Command line + - ["python", "countbolt.py"] + # Output field(s) + - ["word","count"] + parallelism: 1 + + # Logging + - id: "log" + className: "org.apache.storm.flux.wrappers.bolts.LogInfoBolt" + parallelism: 1 + +# Stream definitions +streams: + - name: "Spout --> Splitter" # name isn't used (placeholder for logging, UI, etc.) + # The stream emitter + from: "sentence-spout" + # The stream consumer + to: "splitter-bolt" + # Grouping type + grouping: + type: SHUFFLE + + - name: "Splitter -> Counter" + from: "splitter-bolt" + to: "counter-bolt" + grouping: + type: FIELDS + # field(s) to group on + args: ["word"] + + - name: "Counter -> Log" + from: "counter-bolt" + to: "log" + grouping: + type: SHUFFLE
\ No newline at end of file |