ReadBenchmark.java

import java.io.BufferedReader;
import java.io.FileReader;
import java.io.IOException;
import java.util.Arrays;

import org.apache.commons.math.stat.descriptive.DescriptiveStatistics;

import com.google.common.base.CharMatcher;
import com.mongodb.BasicDBObject;
import com.mongodb.DB;
import com.mongodb.DBCollection;
import com.mongodb.DBCursor;
import com.mongodb.DBObject;
import com.mongodb.Mongo;
import com.mongodb.MongoOptions;
import com.mongodb.ServerAddress;

public class ReadBenchmark {
	private static CharMatcher stripQuotes = CharMatcher.is('"');
	
	public static void main(String[] args) throws IOException {
		// Connect to mongoDB
		MongoOptions mo = new MongoOptions();
		mo.autoConnectRetry = true;
		
		Mongo mongo = new Mongo(Arrays.asList(new ServerAddress("data-node-1"), new ServerAddress("data-node-2"), new ServerAddress("data-node-3")), mo);
		DB db = mongo.getDB("datamining");
		db.setReadOnly(true);
		
		DBCollection docs = db.getCollection("canonical_documents");
		
		for (int i=0; i<10000; i++) {
			// Create new stats collector
			DescriptiveStatistics stats = new DescriptiveStatistics();
			
			// Run though gets
			BufferedReader r = new BufferedReader(new FileReader(args[0]));
			for (String id = r.readLine(); id != null; id = r.readLine()) {
				try {
					// Strip quotes
					id = stripQuotes.trimFrom(id);
					
					// Do stats on response time
					DBObject key = new BasicDBObject();
					key.put("id", id);
					
					// Benchmark this
					long time = System.nanoTime();
					DBCursor results = docs.find(key);
					
					// Get the results
					int n = 0;
					for (DBObject o: results) {
						System.out.println(o);
						n++;
					}
					results.close();
					long delta = System.nanoTime() - time;
					
					stats.addValue(delta);
				} catch (RuntimeException e) {
					System.out.println(e);
					continue;
				}
			}
			
			// Dump out stats
			System.out.println(i + ":" + stats.getMean()/1000 + ", " + stats.getPercentile(99.00)/1000);
		}
	}
}