Github user greghogan commented on a diff in the pull request:
https://github.com/apache/flink/pull/5277#discussion_r160794027
--- Diff: docs/concepts/runtime.md ---
@@ -28,12 +28,12 @@ under the License.
## Tasks and Operator Chains
-For distributed execution, Flink *chains* operator subtasks together into
*tasks*. Each task is executed by one thread.
+For distributed execution, Flink *chains* operator subtasks together into
*tasks*, with one thread executing each task.
Chaining operators together into tasks is a useful optimization: it
reduces the overhead of thread-to-thread
-handover and buffering, and increases overall throughput while decreasing
latency.
-The chaining behavior can be configured; see the [chaining
docs](../dev/datastream_api.html#task-chaining-and-resource-groups) for details.
+handover and buffering and increases overall throughput while decreasing
latency.
+You can configure the chaining behavior, read the [chaining
docs](../dev/datastream_api.html#task-chaining-and-resource-groups) for details.
-The sample dataflow in the figure below is executed with five subtasks,
and hence with five parallel threads.
+Five subtasks execute the sample data flow in the figure below with five
parallel threads.
--- End diff --
The community as well as other projects refer to "dataflow" extensively
(also below).
---
