update
diff --git a/examples/diningphils/index.html b/examples/diningphils/index.html
index 4a39998..3408b9f 100644
--- a/examples/diningphils/index.html
+++ b/examples/diningphils/index.html
@@ -5,6 +5,8 @@
We use the simple dining philosophers example here to give an intuition about the rule-based modeling in Henshin and to conduct a benchmark for the state space generation tool.
</p>
+<h3>Transformation System</h3>
+
<p>
The transformation consists of four rules shown below.
</p>
@@ -17,59 +19,13 @@
In rule <i>left(p)</i> philosopher <i>p</i> picks up his left fork, and analogously for the rule <i>right(p)</i>. In rule <i>release(p)</i> the philospher <i>p</i> releases his fork again. The most interesting rule is <i>createPhilosopher(x)</i> where <i>x</i> refers to the highest philosopher ID in the model (note that Henshin can find all parameters automatically here). The rule <i>createPhilosopher(x)</i> can be used to dynamically add a philosopher to the table and to link it to its neighbors. We use this rule to derive initial models for different numbers of philosophers below.
</p>
-<pre class="sh_java">
- // Create a resource set with a base directory:
- StateSpaceResourceSet resourceSet = new StateSpaceResourceSet(
- "src/org/eclipse/emf/henshin/examples/diningphils/model");
-
- // Load the state space and create a state space manager:
- StateSpace stateSpace = resourceSet.getStateSpace("3-phils.statespace");
- StateSpaceManager manager = StateSpaceFactory.eINSTANCE.createStateSpaceManager(stateSpace);
-
- // To improve the performance, we omit the identity types:
- stateSpace.getProperties().remove(StateSpaceProperties.IDENTITY_TYPES);
-
- // Find the rule for adding a philosopher:
- Rule createPhilRule = stateSpace.getRules().get(0).
- getTransformationSystem().findRuleByName("createPhil");
+<h3>Benchmark</h3>
- // Now do the benchmark...
- try {
- for (int phils=3; true; phils++) {
-
- // First reset the state space:
- manager.resetStateSpace();
- System.gc();
- System.gc();
- System.gc();
-
- // Then explore it again:
- long time = System.currentTimeMillis();
- int expectedStates = (int) Math.pow(3, phils);
- StateSpaceExplorationHelper.doExploration(manager, expectedStates, new NullProgressMonitor());
- if (stateSpace.getStateCount()!=expectedStates || !stateSpace.getOpenStates().isEmpty()) {
- throw new StateSpaceException("Unexpected number of states");
- }
- time = (System.currentTimeMillis() - time);
-
- System.out.println("Philosophers: " + phils);
- System.out.println("States: " + stateSpace.getStateCount());
- System.out.println("Transitions: " + stateSpace.getTransitionCount());
- System.out.println("Time: " + time + "ms");
- System.out.println();
-
- // Add a philosopher:
- EmfGraph initialStateGraph = manager.getModel(stateSpace.getInitialStates().get(0)).getEmfGraph();
- EmfEngine engine = new EmfEngine(initialStateGraph);
- RuleApplication app = new RuleApplication(engine, createPhilRule);
- app.apply();
-
- }
-
- } catch (StateSpaceException e) {
- e.printStackTrace();
- }
-</pre>
+<p>
+We conducted a benchmark to measure the speed of the state space generator.
+The source code for the benchmark can be found
+<a href="http://dev.eclipse.org/svnroot/modeling/org.eclipse.emft.henshin/trunk/plugins/org.eclipse.emf.henshin.examples/src/org/eclipse/emf/henshin/examples/diningphils/DiningPhilsBenchmark.java">here</a>.
+</p>
<p>
The following benchmark was conducted on a Intel(R) Xeon(R) CPU @ 2.50GHz with 4 cores and 8GB of main memory.
