onewaterdrop

public class BlockingQueue implements Queue {
 
    private java.util.Queue queue = new java.util.LinkedList();
 
    /**
     * Make a blocking Dequeue call so that we'll only return when the queue has
     * something on it, otherwise we'll wait until something is put on it.
     * 
     * @returns  This will return null if the thread wait() call is interrupted.
     */

onewaterdrop

package foo;

import edu.stanford.nlp.fsm.ExactGrammarCompactor;
import edu.stanford.nlp.io.IOUtils;
import edu.stanford.nlp.io.NumberRangeFileFilter;
import edu.stanford.nlp.io.NumberRangesFileFilter;
import edu.stanford.nlp.ling.*;
import edu.stanford.nlp.objectbank.TokenizerFactory;
import edu.stanford.nlp.parser.ViterbiParser;
import edu.stanford.nlp.parser.KBestViterbiParser;

onewaterdrop

System Design Cheatsheet

Picking the right architecture = Picking the right battles + Managing trade-offs

Basic Steps

1. Clarify and agree on the scope of the system

• User cases (description of sequences of events that, taken together, lead to a system doing something useful)
  • Who is going to use it?
  • How are they going to use it?

onewaterdrop

Latency Comparison Numbers
--------------------------
L1 cache reference                           0.5 ns
Branch mispredict                            5   ns
L2 cache reference                           7   ns                      14x L1 cache
Mutex lock/unlock                           25   ns
Main memory reference                      100   ns                      20x L2 cache, 200x L1 cache
Compress 1K bytes with Zippy             3,000   ns        3 us
Send 1K bytes over 1 Gbps network       10,000   ns       10 us
Read 4K randomly from SSD*             150,000   ns      150 us          ~1GB/sec SSD