Update 10-further-mpi-topics.md

mpi/docs/10-further-mpi-topics.md

@@ -168,63 +168,6 @@ MPI_Win_free(&window);
 ```
 
 
-# Limitations for data access
-
-- Compatibility of local and remote operations when multiple processes
-  access a window during an epoch
-
-![](img/one-sided-limitations.png)
-
-
-# Advanced synchronization:
-
-- Assert argument in `MPI_Win_fence`:
-
-    `MPI_MODE_NOSTORE`
-    : The local window was not updated by local stores (or local get or
-      receive calls) since last synchronization
-
-    `MPI_MODE_NOPUT`
-    : The local window will not be updated by put or accumulate calls after
-      the fence call, until the ensuing (fence) synchronization
-
-    `MPI_MODE_NOPRECEDE`
-    : The fence does not complete any sequence of locally issued RMA calls
-
-    `MPI_MODE_NOSUCCEED`
-    : The fence does not start any sequence of locally issued RMA calls
-
-
-# Advanced synchronization
-
-- More control on epochs can be obtained by starting and ending the
-  exposure and access epochs separately
-- Target: Exposure epoch
-    - Start: `MPI_Win_post`
-    - End: `MPI_Win_wait`
-- Origin: Access epoch
-    - Start: `MPI_Win_start`
-    - End: `MPI_Win_complete`
-
-
-# Enhancements in MPI-3
-
-- New window creation function: `MPI_Win_allocate`
-    - Allocate memory and create window at the same time
-- Dynamic windows: `MPI_Win_create_dynamic`, `MPI_Win_attach`,
-  `MPI_Win_detach`
-  - Non-collective exposure of memory
-
-
-# Enhancements in MPI-3
-
-- New data movement operations: `MPI_Get_accumulate`, `MPI_Fetch_and_op`,
-  `MPI_Compare_and_swap`
-- New memory model `MPI_Win_allocate_shared`
-    - Allocate memory which is shared between MPI tasks
-- Enhancements for passive target synchronization
-
-
 # Performance considerations
 
 - Performance of the one-sided approach is highly implementation-dependent