Merge pull request #5972 from aravindksg/ofi_sep_master

MTL/OFI: Add OFI Scalable Endpoint support

Author: matcabral

ompi/mca/mtl/ofi/README

@@ -1,5 +1,5 @@
-OFI MTL
-
+OFI MTL:
+--------
 The OFI MTL supports Libfabric (a.k.a. Open Fabrics Interfaces OFI,
 https://ofiwg.github.io/libfabric/) tagged APIs (fi_tagged(3)). At
 initialization time, the MTL queries libfabric for providers supporting tag matching
@@ -9,19 +9,22 @@ The user may modify the OFI provider selection with mca parameters
 mtl_ofi_provider_include or mtl_ofi_provider_exclude.
 
 PROGRESS:
+---------
 The MTL registers a progress function to opal_progress. There is currently
 no support for asynchronous progress. The progress function reads multiple events
 from the OFI provider Completion Queue (CQ) per iteration (defaults to 100, can be
 modified with the mca mtl_ofi_progress_event_cnt) and iterates until the
 completion queue is drained.
 
 COMPLETIONS:
+------------
 Each operation uses a request type ompi_mtl_ofi_request_t which includes a reference
-to an operation specific completion callback, an MPI request, and a context.  The
+to an operation specific completion callback, an MPI request, and a context. The
 context (fi_context) is used to map completion events with MPI_requests when reading the
 CQ.
 
 OFI TAG:
+--------
 MPI needs to send 96 bits of information per message (32 bits communicator id,
 32 bits source rank, 32 bits MPI tag) but OFI only offers 64 bits tags. In
 addition, the OFI MTL uses 2 bits of the OFI tag for the synchronous send protocol.
@@ -67,3 +70,76 @@ This is signaled in mem_tag_format (see fi_endpoint(3)) by setting higher order
 to zero. In such cases, the OFI MTL will reduce the number of communicator ids supported
 by reducing the bits available for the communicator ID field in the OFI tag.
 
+SCALABLE ENDPOINTS:
+-------------------
+OFI MTL supports OFI Scalable Endpoints feature as a means to improve
+multi-threaded application throughput and message rate. Currently the feature
+is designed to utilize multiple TX/RX contexts exposed by the OFI provider in
+conjunction with a multi-communicator MPI application model. Therefore, new OFI
+contexts are created as and when communicators are duplicated in a lazy fashion
+instead of creating them all at once during init time and this approach also
+favours only creating as many contexts as needed.
+
+1. Multi-communicator model:
+   With this approach, the application first duplicates the communicators it
+   wants to use with MPI operations (ideally creating as many communicators as
+   the number of threads it wants to use to call into MPI). The duplicated
+   communicators are then used by the corresponding threads to perform MPI
+   operations. A possible usage scenario could be in an MPI + OMP
+   application as follows (example limited to 2 ranks):
+
+    MPI_Comm dup_comm[n];
+    MPI_Init_thread(&argc, &argv, MPI_THREAD_MULTIPLE, &provided);
+     for (i = 0; i < n; i++) {
+        MPI_Comm_dup(MPI_COMM_WORLD, &dup_comm[i]);
+     }
+     if (rank == 0) {
+#pragma omp parallel for private(host_sbuf, host_rbuf) num_threads(n)
+         for (i = 0; i < n ; i++) {
+                    MPI_Send(host_sbuf, MYBUFSIZE, MPI_CHAR,
+                                        1, MSG_TAG, dup_comm[i]);
+                    MPI_Recv(host_rbuf, MYBUFSIZE, MPI_CHAR,
+                                       1, MSG_TAG, dup_comm[i], &status);
+         }
+    } else if (rank == 1) {
+#pragma omp parallel for private(status, host_sbuf, host_rbuf) num_threads(n)
+            for (i = 0; i < n ; i++) {
+                MPI_Recv(host_rbuf, MYBUFSIZE, MPI_CHAR,
+                                   0, MSG_TAG, dup_comm[i], &status);
+                MPI_Send(host_sbuf, MYBUFSIZE, MPI_CHAR,
+                                   0, MSG_TAG, dup_comm[i]);
+           }
+    }
+
+2. MCA variable:
+To utilize the feature, the following MCA variable needs to be set:
+  mtl_ofi_thread_grouping:
+  This MCA variable is at the OFI MTL level and needs to be set to switch
+  the feature on.
+
+  Default: 0
+
+  It is not recommended to set the MCA variable for:
+   - Multi-threaded MPI applications not following multi-communicator approach.
+   - Applications that have multiple threads using a single communicator as
+     it may degrade performance.
+
+Command-line syntax to set the MCA variable:
+  "-mca mtl_ofi_thread_grouping 1"
+
+3. Notes on performance:
+  - OFI MTL will create as many TX/RX contexts as allowed by an underlying
+    provider (each provider may have different thresholds). Once the threshold
+    is exceeded, contexts are used in a round-robin fashion which leads to
+    resource sharing among threads. Therefore locks are required to guard
+    against race conditions. For performance, it is recommended to have
+
+      Number of communicators = Number of contexts
+
+    For example, when using PSM2 provider, the number of contexts is dictated
+    by the Intel Omni-Path HFI1 driver module.
+
+  - For applications using a single thread with multiple communicators and MCA
+    variable "mtl_ofi_thread_grouping" set to 1, the MTL will use multiple
+    contexts, but the benefits may be negligible as only one thread is driving
+    progress.

ompi/mca/mtl/ofi/help-mtl-ofi.txt

@@ -24,3 +24,19 @@ fi_info -v -p %s
 
   Local host: %s
   Location: %s:%d
+
+[SEP unavailable]
+Scalable Endpoint feature is required for Thread Grouping feature to work
+but it is not supported by %s provider. Try disabling this feature.
+
+  Local host: %s
+  Location: %s:%d
+
+[SEP ctxt limit]
+Reached limit (%d) for number of OFI contexts that can be opened with the
+provider. Creating new communicators beyond this limit is possible but
+they will re-use existing contexts in round-robin fashion.
+Using new communicators beyond the limit will impact performance.
+
+  Local host: %s
+  Location: %s:%d