Data pipeline: condense input and output

Signed-off-by: Lynette Miles <[email protected]>

Author: esmerel

concepts/data-pipeline.md

@@ -9,6 +9,8 @@ The [`buffer`](./buffering.md) phase in the pipeline aims to provide a unified a
 ## Filter
 
 In production environments you need full control of the data you're collecting. Filtering lets you alter the collected data before delivering it to a destination.
-For more details about the Filters available and their usage, see [Filters](https://docs.fluentbit.io/manual/pipeline/filters).
+For more details about the Filters available and their usage, see [Filters](../pipeline/filters.md).
 
 ## Inputs
+
+Fluent Bit provides [input plugins](../pipeline/inputs.md) to gather information from different sources. Some plugins collect data from log files, while others can gather metrics information from the operating system. There are many plugins to suit different needs.

concepts/data-pipeline/output.md

@@ -1,10 +1,6 @@
----
-description: The way to gather data from your sources
----
+# Inputs
 
-# Input
-
-[Fluent Bit](http://fluentbit.io) provides input plugins to gather information from different sources. Some plugins collect data from log files, while others can gather metrics information from the operating system. There are many plugins to suit different needs.
+Input plugins gather information from different sources. Some plugins collect data from log files, while others can gather metrics information from the operating system. There are many plugins to suit different needs.
 
 ```mermaid
 graph LR
@@ -21,7 +17,3 @@ graph LR
 ```
 
 When an input plugin loads, an internal _instance_ is created. Each instance has its own independent configuration. Configuration keys are often called _properties_.
-
-Every input plugin has its own documentation section that specifies how to use it and what properties are available.
-
-For more details, see [Input Plugins](https://docs.fluentbit.io/manual/pipeline/inputs).