Remove Turtle, Graphalytics, and old SNB-related code

.gitignore

@@ -54,3 +54,4 @@ datagen_output/
 
 # configuration file
 params.ini
+dependency-reduced-pom.xml

src/main/java/ldbc/snb/datagen/DatagenMode.java

@@ -3,6 +3,5 @@
 public enum DatagenMode {
     INTERACTIVE,
     BI,
-    GRAPHALYTICS,
     RAW_DATA
 }

src/main/java/ldbc/snb/datagen/DatagenParams.java

@@ -329,12 +329,6 @@ public static void readConf(LdbcConfiguration conf) {
             dateFormat = conf.get("generator.StringDate.dateFormat");
             endYear = startYear + numYears;
 
-            if (conf.get("ldbc.snb.datagen.generator.gscale") != null) {
-                double scale = conf.getDouble("ldbc.snb.datagen.generator.gscale", 6.0);
-                DegreeDistribution degreeDistribution = getDegreeDistribution();
-                degreeDistribution.initialize(conf);
-                numPersons = findNumPersonsFromGraphalyticsScale(degreeDistribution, scale);
-            }
             System.out.println(" ... Datagen Mode " + datagenMode);
             System.out.println(" ... Num Persons " + numPersons);
             System.out.println(" ... Start Year " + startYear);
@@ -357,9 +351,6 @@ public static DatagenMode getDatagenMode() {
             case "bi":
                 mode = BI;
                 break;
-            case "graphalytics":
-                mode = GRAPHALYTICS;
-                break;
             case "rawdata":
                 mode = RAW_DATA;
                 break;
@@ -375,31 +366,6 @@ private static double scale(long numPersons, double mean) {
         return Math.log10(mean * numPersons / 2 + numPersons);
     }
 
-    private static long findNumPersonsFromGraphalyticsScale(DegreeDistribution distribution, double scale) {
-
-        long numPersonsMin = 1000000;
-        while (scale(numPersonsMin, distribution.mean(numPersonsMin)) > scale)
-            numPersonsMin /= 2;
-
-        long numPersonsMax = 1000000;
-        while (scale(numPersonsMax, distribution.mean(numPersonsMax)) < scale)
-            numPersonsMax *= 2;
-
-        long currentNumPersons = (numPersonsMax - numPersonsMin) / 2 + numPersonsMin;
-        double currentScale = scale(currentNumPersons, distribution.mean(currentNumPersons));
-        while (Math.abs(currentScale - scale) / scale > 0.001) {
-            if (currentScale < scale)
-                numPersonsMin = currentNumPersons;
-            else
-                numPersonsMax = currentNumPersons;
-
-            currentNumPersons = (numPersonsMax - numPersonsMin) / 2 + numPersonsMin;
-            currentScale = scale(currentNumPersons, distribution.mean(currentNumPersons));
-            System.out.println(numPersonsMin + " " + numPersonsMax + " " + currentNumPersons + " " + currentScale);
-        }
-        return currentNumPersons;
-    }
-
     //TODO: add remaining degree distributions
     public static DegreeDistribution getDegreeDistribution() {
 

src/main/java/ldbc/snb/datagen/hadoop/LdbcDatagen.java

@@ -114,16 +114,14 @@ private void copyToLocal(FileSystem fs, String pathString) throws Exception {
 
     private long personActivityJob(String hadoopPrefix, Configuration conf, FileSystem fs) throws Exception {
         long startPersonActivity = System.currentTimeMillis();
-        if (DatagenParams.getDatagenMode() != GRAPHALYTICS) {
-            printProgress("Generating and serializing person activity");
-            HadoopPersonActivityGenerator activityGenerator = new HadoopPersonActivityGenerator(conf);
-            activityGenerator.run(hadoopPrefix + "/mergedPersons");
-            for (int i = 0; i < HadoopConfiguration.getNumThreads(conf); ++i) {
-                if (i < (int) Math.ceil(DatagenParams.numPersons / (double) DatagenParams.blockSize)) { // i<number of blocks
-                    copyToLocal(fs, HadoopConfiguration.getHadoopDir(conf) + "/m" + i + "personFactors.txt");
-                    copyToLocal(fs, HadoopConfiguration.getHadoopDir(conf) + "/m" + i + "activityFactors.txt");
-                    copyToLocal(fs, HadoopConfiguration.getHadoopDir(conf) + "/m0friendList" + i + ".csv");
-                }
+        printProgress("Generating and serializing person activity");
+        HadoopPersonActivityGenerator activityGenerator = new HadoopPersonActivityGenerator(conf);
+        activityGenerator.run(hadoopPrefix + "/mergedPersons");
+        for (int i = 0; i < HadoopConfiguration.getNumThreads(conf); ++i) {
+            if (i < (int) Math.ceil(DatagenParams.numPersons / (double) DatagenParams.blockSize)) { // i<number of blocks
+                copyToLocal(fs, HadoopConfiguration.getHadoopDir(conf) + "/m" + i + "personFactors.txt");
+                copyToLocal(fs, HadoopConfiguration.getHadoopDir(conf) + "/m" + i + "activityFactors.txt");
+                copyToLocal(fs, HadoopConfiguration.getHadoopDir(conf) + "/m0friendList" + i + ".csv");
             }
         }
         return (System.currentTimeMillis() - startPersonActivity);

src/main/java/ldbc/snb/datagen/hadoop/serializer/HadoopPersonSortAndSerializer.java

@@ -104,7 +104,7 @@ public void reduce(BlockKey key, Iterable<Person> valueSet, Context context) thr
 
             for (Person p : valueSet) {
 
-                if (DatagenParams.getDatagenMode() == DatagenMode.RAW_DATA || DatagenParams.getDatagenMode() == DatagenMode.GRAPHALYTICS) {
+                if (DatagenParams.getDatagenMode() == DatagenMode.RAW_DATA) {
                     dynamicPersonSerializer.export(p);
                     for (Knows k : p.getKnows()) {
                         dynamicPersonSerializer.export(p, k);