docs: update examples (numaproj#40)

Signed-off-by: Edward Lee <[email protected]>

Author: edlee2121

README.md

@@ -24,6 +24,7 @@ NumaFlow installs in less than a minute and is easier and cheaper for simple dat
 * Data aggregation (e.g. group-by)
 
 ## Resources
-- Check out [QUICK START](docs/QUICK_START.md) to try it out.
-- Take a look at [DEVELOPMENT](docs/DEVELOPMENT.md) to set up development environment.
-- Refer to [CONTRIBUTING](https://github.com/numaproj/numaproj/blob/main/CONTRIBUTING.md) to contribute to the project.
+- [QUICK_START](docs/QUICK_START.md)
+- [EXAMPLES](examples)
+- [DEVELOPMENT](docs/DEVELOPMENT.md)
+- [CONTRIBUTING](https://github.com/numaproj/numaproj/blob/main/CONTRIBUTING.md)

examples/1-simple-pipeline.yaml

@@ -4,7 +4,7 @@ metadata:
   name: simple-pipeline
 spec:
   vertices:
-    - name: input
+    - name: in
       source:
         # A self data generating source
         generator:
@@ -14,12 +14,12 @@ spec:
       udf:
         builtin:
           name: cat # A builtin UDF which simply cats the message
-    - name: output
+    - name: out
       sink:
         # A simple log printing sink
         log: {}
   edges:
-    - from: input
+    - from: in
       to: cat
     - from: cat
-      to: output
+      to: out

examples/2-http-pipeline.yaml

@@ -4,7 +4,7 @@ metadata:
   name: http-pipeline
 spec:
   vertices:
-    - name: input
+    - name: in
       source:
         # An http endpoint
         http:
@@ -21,12 +21,12 @@ spec:
       udf:
         builtin:
           name: cat # A builtin UDF which simply cats the message
-    - name: output
+    - name: out
       sink:
         # A simple log printing sink
         log: {}
   edges:
-    - from: input
+    - from: in
       to: cat
     - from: cat
-      to: output
+      to: out

examples/3-pipeline-filter.yaml

@@ -4,7 +4,7 @@ metadata:
   name: filter-pipeline
 spec:
   vertices:
-    - name: input
+    - name: in
       source:
         # A self data generating source
         http:
@@ -22,12 +22,12 @@ spec:
           name: filter
           kwargs:
             expression: int(json(payload).id) < 100
-    - name: output
+    - name: out
       sink:
         # A simple log printing sink
         log: {}
   edges:
-    - from: input
+    - from: in
       to: filter
     - from: filter
-      to: output
+      to: out

examples/4-udsink-pipeline.yaml

@@ -4,7 +4,7 @@ metadata:
   name: udsink-pipeline
 spec:
   vertices:
-    - name: input
+    - name: in
       source:
         generator:
           rpu: 1
@@ -14,13 +14,13 @@ spec:
       udf:
         builtin:
           name: cat
-    - name: output
+    - name: out
       sink:
         udsink:
           container:
             image: my-sink:latest
   edges:
-    - from: input
+    - from: in
       to: p1
     - from: p1
-      to: output
+      to: out

examples/5-auto-scaling.yaml

@@ -4,7 +4,7 @@ metadata:
   name: auto-scaling-pipeline
 spec:
   vertices:
-    - name: input
+    - name: in
       source:
         # A self data generating source
         generator:
@@ -17,15 +17,15 @@ spec:
       udf:
         builtin:
           name: cat # A builtin UDF which simply cats the message
-    - name: output
+    - name: out
       scale:
         min: 3 # Minimal replicas
         max: 5 # Maximum replicas
       sink:
         # A simple log printing sink
         log: {}
   edges:
-    - from: input
+    - from: in
       to: cat
     - from: cat
-      to: output
+      to: out

examples/README.md

@@ -1,18 +1,18 @@
 # NumaFlow by Example
 
-Welcome to the NumaFlow Community! This document provides an example-by example guide to using NumaFlow.
+Welcome to the NumaFlow Community! This document provides an example-by-example guide to using NumaFlow.
 
 If you haven't already, install NumaFlow by following the [QUICK START](../docs/QUICK_START.md) instructions.
 
-The top-level abstraction in NumaFlow is the `Pipeline`. A `Pipeline` consists of a set of `vertices` connected by `edges`. A vertex can be a `source`, `sink`, or `processing` vertex. In the example below, we have a source vertex named *input* that generates messages at a specified rate, a sink vertex named *ouput* that logs messages, and a processing vertex named *cat* that produces any input message as output. Lastly, there are two edges, one connecting the *input* to the *cat* vertex and another connecting the *cat* to the *output* vertex. The resulting pipeline simply copies internally generated messages to the log.
+The top-level abstraction in NumaFlow is the `Pipeline`. A `Pipeline` consists of a set of `vertices` connected by `edges`. A vertex can be a `source`, `sink`, or `processing` vertex. In the example below, we have a source vertex named *in* that generates messages at a specified rate, a sink vertex named *out* that logs messages, and a processing vertex named *cat* that produces any input message as output. Lastly, there are two edges, one connecting the *in* to the *cat* vertex and another connecting the *cat* to the *out* vertex. The resulting pipeline simply copies internally generated messages to the log.
 ```yaml
 apiVersion: numaflow.numaproj.io/v1alpha1
 kind: Pipeline
 metadata:
   name: simple-pipeline
 spec:
   vertices:
-    - name: input
+    - name: in
       source:
         # Generate 5 messages every second
         # xxx what does rpu stand for? how large are the messages by default, what data is contained in the message? rename duration to interval?
@@ -23,18 +23,18 @@ spec:
       udf: # A user defined function
         builtin: # Use a builtin function as the udf
           name: cat # cats the message
-    - name: output
+    - name: out
       sink:
         # Output message to the log
 	# xxx which log? k8s events?
         log: {}
 
-  # input -> cat -> output
+  # in -> cat -> out
   edges:
-    - from: input
+    - from: in
       to: cat
     - from: cat
-      to: output
+      to: out
 ```
 
 Below we have a simple variation on the above example that takes input from an http endpoint.
@@ -45,7 +45,7 @@ metadata:
   name: http-pipeline
 spec:
   vertices:
-    - name: input
+    - name: in
       source:
         http:
           # Whether to create a ClusterIP Service, defaults to false
@@ -62,26 +62,26 @@ spec:
       udf:
         builtin:
           name: cat # A builtin UDF which simply cats the message
-    - name: output
+    - name: out
       sink:
         # A simple log printing sink
         log: {}
   edges:
-    - from: input
+    - from: in
       to: cat
     - from: cat
-      to: output
+      to: out
 ```
 
-Let's modify the UDF in the first example to passthrough only messages with an *id* less than 100
+Let's modify the UDF in the first example to pass-through only messages with an *id* less than 100
 ```yaml
 apiVersion: numaflow.numaproj.io/v1alpha1
 kind: Pipeline
 metadata:
   name: filter-pipeline
 spec:
   vertices:
-    - name: input
+    - name: in
       source:
         generator:
           rpu: 5
@@ -92,13 +92,13 @@ spec:
           name: filter
           kwargs:
             expression: int(json(payload).id) < 100
-    - name: output
+    - name: out
       sink:
         log: {}
   edges:
-    - from: input
+    - from: in
       to: filter
     - from: filter
-      to: output
+      to: out
 ```