Add offline HTML dashboards for sim stats and HW correlation plots

README.md

@@ -182,6 +182,17 @@ After the jobs finish - you can collect all the stats using:
 ./util/job_launching/get_stats.py -N myTest | tee stats.csv
 ```
 
+To plot them, either write one HTML file per stat:
+```bash
+./util/plotting/plot-get-stats.py -c stats.csv
+# outputs under ./util/plotting/htmls/
+```
+or build a single offline dashboard (searchable sidebar, one chart at a time):
+```bash
+./util/plotting/plot-get-stats-dashboard.py -c stats.csv
+# open ./util/plotting/htmls/dashboard.html
+```
+
 If you want to run the accel-sim.out executable command itself for specific workload, you can use:
 ```bash
 /gpu-simulator/bin/release/accel-sim.out -trace ./hw_run/rodinia_2.0-ft/9.1/backprop-rodinia-2.0-ft/4096___data_result_4096_txt/traces/kernelslist.g -config ./gpu-simulator/gpgpu-sim/configs/tested-cfgs/SM7_QV100/gpgpusim.config -config ./gpu-simulator/configs/tested-cfgs/SM7_QV100/trace.config
@@ -230,11 +241,11 @@ To run the correlator - do the following:
 ```
 
 The script may take a few minutes to run (primarily because it is parsing a large amount of hardware data for >150 apps).
-Stdout will print the summary of counters error, correlation, etc. and a set of correlation plots will be generated
-in:
+Stdout will print the summary of counters error, correlation, etc. By default it writes one offline dashboard:
 ```
-./util/plotting/correl-html/
+./util/plotting/correl-html/dashboard.html
 ```
+Use `--individual` for the old per-stat HTML files, or `--all-html` for both.
 
 Here you will find interactive HTML plots, csvs and textual summaries of how well the simulator correlated against hardware on both a per-kernel and per-app basis.
 Note that the simple tests we ran in this tutorial are short running and not generally representative of scaled GPU apps and are just meant to quickly validate you can get Accel-Sim working.

util/plotting/README.md

@@ -23,6 +23,31 @@
 All stats collected by the `get_stats.py` file will be plotted and placed in ./htmls/.
 [An example for the IPC is here](https://engineering.purdue.edu/tgrogers/accel-sim/example-plots/example.plot.rodinia_2.0-ft.html).
 
+# Single-page dashboard (all stats in one HTML)
+
+`./plot-get-stats.py` writes one HTML file per stat (~25 files). If you'd rather
+review everything in one place, use `./plot-get-stats-dashboard.py`, which reads
+the exact same CSV and produces a single, self-contained `dashboard.html` with a
+searchable sidebar. Only one stat's chart is shown at a time; click a stat in the
+sidebar to switch. Stat names are cleaned up from the collection regexes (e.g.
+`gpgpu_simulation_time\s*=.*\(([0-9]+) sec\).*` becomes `gpgpu_simulation_time (sec)`),
+and Plotly is embedded so the file opens in any browser with no internet.
+
+```bash
+# Collect the stats (same as above):
+../job_launching/get_stats.py -R -C QV100-SASS,QV100-PTX -B rodinia_2.0-ft | tee per-app-stats.csv
+
+# Build the dashboard:
+./plot-get-stats-dashboard.py -c per-app-stats.csv
+# open ./htmls/dashboard.html
+```
+
+Options:
+* `-c/--csv_file`  the get_stats.py CSV to plot (required)
+* `-o/--output`    output HTML path (default `./htmls/dashboard.html`)
+* `-n/--basename`  dashboard title
+* `-s/--stats_yml` stats yml used to group the sidebar into categories (defaults to `../job_launching/stats/example_stats.yml`)
+
 # Instructions on plotting correlation graphs
 
 * Make sure CUDA\_INSTALL\_PATH is set and bin/lib directories are in PATH and LD\_LIBRARY\_PATH
@@ -40,11 +65,18 @@ All stats collected by the `get_stats.py` file will be plotted and placed in ./h
     ../job_launching/get_stats.py -R -K -k -C <Your config name> -B <simulator apps> > correl.stats.csv
     # An example: ../job_launching/get_stats.py -R -K -k -C QV100-SASS,QV100-PTX -B rodinia_2.0-ft > correl.stats.csv
     ./plot-correlation.py -c correl.stats.csv
-    # stdout will print summary statistics and html files will be generated in ./correl-html/
+    # Default: one offline dashboard at ./correl-html/dashboard.html
+    #   (sidebar + Per-app/Per-kernel toggle; no per-stat HTML spam)
+    # Per-stat HTML files only (old behavior, no dashboard):
+    #   ./plot-correlation.py -c correl.stats.csv -H ... --individual
+    # Both dashboard and per-stat HTML:
+    #   ./plot-correlation.py -c correl.stats.csv -H ... --all-html
     # You can generate pdf files instead using
     ./plot-correlation.py -c correl.stats.csv -H ../../hw_run/QUADRO-V100/9.1/
     # You can also generate pdf files for the correaltions using "-i pdf"
     ```
+For day-to-day review, prefer the default `./correl-html/dashboard.html`. Use `--individual`
+only when you need the separate per-app / per-kernel HTML files for archival or sharing.
 [Here is an example correlation plot for the simple rodinia tests aggregated per-app](https://engineering.purdue.edu/tgrogers/accel-sim/example-plots/gv100-cycles.QV100-PTX.QV100-SASS.per-app.html).
 [And per-kernel](https://engineering.purdue.edu/tgrogers/accel-sim/example-plots/gv100-cycles.QV100-PTX.QV100-SASS.per-kernel.html).
 Note again - that these short-running tests are not representative of longer running GPU apps and the correlation on these applications should