Add model stats to Ping Pong example

doc/codes-vis-readme.md

@@ -6,6 +6,8 @@ on the ROSS webpage.
 
 There are currently 4 types of instrumentation: GVT-based, real time sampling, virtual time sampling, and event tracing.
 See the ROSS documentation for more info on the specific options or use `--help` with your model.
+The binary layout of the model-stats output files (and a reference Python parser) is documented in
+[model-stats-binary-format.md](model-stats-binary-format.md).
 To collect data about the simulation engine, no changes are needed to model code for any of the instrumentation modes.
 Some additions to the model code is needed in order to turn on any model-level data collection.
 See the "Model-level data sampling" section on [ROSS Instrumentation blog post](http://ross-org.github.io/instrumentation/instrumentation.html).
@@ -109,7 +111,7 @@ st_model_types svr_model_types[] = {
 
 static void svr_register_model_types()
 {
-    st_model_type_register("ns-lp", &svr_model_types[0]);
+    st_model_type_register("nw-lp", &svr_model_types[0]);
 }
 
 int main(int argc, char **argv)
@@ -135,12 +137,18 @@ modes are collecting model-level data as well.
 ### CODES LPs that currently have event type collection implemented:
 If you're using any of the following CODES models, you don't have to add anything, unless you want to change the data that's being collected.
 - nw-lp (model-net-mpi-replay.c)
+- ping pong tutorial server LP (doc/example/tutorial-synthetic-ping-pong.c) — also collects
+  model-level stats in all sampling modes; a good minimal example to copy from
 - dfly server LP (model-net-synthetic.c)
 - custom dfly server LP (model-net-synthetic-custom-dfly.c)
 - fat tree server LP (model-net-synthetic-fattree.c)
 - slimfly server LP (model-net-synthetic-slimfly.c)
 - original dragonfly router and terminal LPs (dragonfly.c)
 - dragonfly custom router and terminal LPs (dragonfly-custom.cxx)
+- dragonfly dally router and terminal LPs (dragonfly-dally.cxx) — richest model-stats
+  payload (packet rates, routing-path mix, stalls, occupancy snapshots); layout in
+  [model-stats-binary-format.md](model-stats-binary-format.md)
+- dragonfly plus router and terminal LPs (dragonfly-plus.cxx)
 - slimfly router and terminal LPs (slimfly.c)
 - fat tree switch and terminal LPs (fat-tree.c)
 - model-net-base-lp (model-net-lp.c)

doc/example/tutorial-ping-pong-surrogate.conf.in

@@ -41,7 +41,7 @@ PARAMS
 # bandwidth in GiB/s for compute node-router channels
    cn_bandwidth="2.0";
 # ROSS message size
-   message_size="440";
+   message_size="456";
 # number of compute nodes connected to router, dictated by dragonfly config
 # file
    num_cns_per_router="2";

doc/example/tutorial-ping-pong-surrogate.yaml.in

@@ -50,7 +50,7 @@ topology:
 # scheduler options (alternative: round-robin)
     modelnet_scheduler: fcfs
 # ROSS message size
-    message_size: 440
+    message_size: 456
 # folder path to store packet latency from terminal to terminal, if no value is given it won't save anything
     save_packet_latency_path: "${PACKET_LATENCY_TRACE_PATH}"
 # folder path to store router-local timing rows for router queueing-delay training

doc/example/tutorial-ping-pong.conf.in

@@ -1,3 +1,7 @@
+# To also collect model-level stats through the ROSS instrumentation layer, add e.g.
+#   --model-stats=1 --num-gvt=8 --stats-path=model-stats-out
+# to the run command (see doc/model-stats-binary-format.md for all modes and the
+# output format; scripts/parse-ross-model-stats.py parses the result).
 LPGROUPS
 {
    MODELNET_GRP
@@ -38,7 +42,7 @@ PARAMS
 # bandwidth in GiB/s for compute node-router channels
    cn_bandwidth="2.0";
 # ROSS message size
-   message_size="440";
+   message_size="456";
 # number of compute nodes connected to router, dictated by dragonfly config
 # file
    num_cns_per_router="2";

doc/example/tutorial-ping-pong.yaml.in

@@ -9,6 +9,11 @@ schema_version: 1
 # as in the .conf. The compiler derives repetitions = num_groups * num_routers and
 # lays out the [workload, terminal, router] LPs, so nw-lp / modelnet_dragonfly_dally*
 # are emitted for you.
+#
+# To also collect model-level stats through the ROSS instrumentation layer, add e.g.
+#   --model-stats=1 --num-gvt=8 --stats-path=model-stats-out
+# to the run command (see doc/model-stats-binary-format.md for all modes and the
+# output format; scripts/parse-ross-model-stats.py parses the result).
 
 components:
   compute_host:
@@ -49,7 +54,7 @@ topology:
 # scheduler options (alternative: round-robin)
     modelnet_scheduler: fcfs
 # ROSS message size
-    message_size: 440
+    message_size: 456
 # router buffer occupancy snapshots
     router_buffer_snapshots: [ ${BUFFER_SNAPSHOTS} ]
 # folder path to store packet latency from terminal to terminal, if no value is given it won't save anything