Update WorkflowRunner.md

Author: sawenzel

MC/doc/WorkflowRunner.md

@@ -13,11 +13,11 @@ not **what** is executed or how it is configured.
 The tool provides features of a typical data/task pipelining environment
 using a DAG approach. It allows to separate the concerns of workflow setup and workflow running - and as such
 allows to optimize workflow deployment during execution.
-It is inspired by similar ideas in ALICE alibuild and O2/DPL but differs in some aspects:
+It is inspired by similar ideas in ALICE alibuild and the ALILE Data Processing Layer (DPL) but differs in some aspects:
 
    * allow to schedule tasks in parallel
-   * be resource aware (e.g., don't schedule 2 tasks requires large memory at the same time)
-   * allow to schedule any task, be at a simple executable, bash script, ROOT macro, or DPL workflow.
+   * be resource aware (e.g., don't schedule 2 tasks requiring large memory at the same time)
+   * allow to schedule any task, be it a simple executable, bash script, ROOT macro, or DPL workflow.
 
 The goals of the tool are:
 
@@ -27,7 +27,8 @@ The goals of the tool are:
    * skip-done features when run again with same input
    * if input changes, rerun only affected stages
    * automatic task skipping in case it is not relevant for the goal
-   * dream: automatic task fusion (say DPL workflow for TPC digitization and TPC reco) when we don't need the intermediate files
+   * file provenance tracking, cleanup of intermediate products
+   * dream: automatic DPL fusion/pipelining (say DPL workflow for TPC digitization and TPC reco) when we don't need the intermediate files
    on disc
 
 # Workflow specification
@@ -71,12 +72,12 @@ Naturally, task2 depends on task1 expressed via the `needs` list.
 Further keys in this format are:
 | field | description |
 | ----- | ----------- |
-| `resources` | estimated resource usage for average cpu load (250 = 2.5 CPUs) and maximal memory in MB. Used for scheduling. |
+| `resources` | estimated resource usage for average cpu load (250 = 2.5 CPUs) and maximal memory in MB. Used for scheduling. -1 is used for unknown or don't care. |
 | `timeframe` | timeframe index or -1 if not associated to any timeframe. May have influence on order of execution (prefer finish timeframe first) |
 | `cwd` | the workding directory where this is to be executed |
 | `label` | a list labels, describing this stage. Can be used to execute workfow in stages (such as 'do all digitization', 'run everthing for ITS' 
 
-A workflow is generated by different tools. A current example following the PWGHF embedding exercise can be found here ..
+A workflow is generated by different tools. A current example following the PWGHF embedding exercise can be found here https://github.com/AliceO2Group/O2DPG/blob/master/MC/run/PWGHF/create_embedding_workflow.py
 
 ## Example usage
 
@@ -88,20 +89,17 @@ o2_dpg_workflow_runner.py -f workflow_sim.json
 
 Show what you would run
 ```
-alienv enter O2/latest O2DPG/latest
 o2_dpg_workflow_runner.py -f workflow_sim.json --dry-run
 ```
 
 ## Future targeted features:
 
 Run until everyting marked "RECO" is done
 ```
-alienv enter O2/latest O2DPG/latest
 o2_dpg_workflow_runner.py -f workflow_sim.json --stages RECO
 ```
 
 Rerun worflow until AOD, skipping all tasks already done
 ```
-alienv enter O2/latest O2DPG/latest
 o2_dpg_workflow_runner.py -f workflow_sim.json --stages AOD --skip-done
 ```