The Computer Language
Benchmarks Game

k-nucleotide Node.js program

source code

/* The Computer Language Benchmarks Game
   http://benchmarksgame.alioth.debian.org/

   Contributed by Jesse Millikan
   Modified by Matt Baker
   Ported, modified, and parallelized by Roman Pletnev
*/

'use strict';

var rd = require('readline'), cluster = require('cluster');

function RefNum(num){ this.num = num; }
RefNum.prototype.toString = function() { return this.num.toString(); }

function frequency(seq, length){
    var freq = new Map(), n = seq.length-length+1, key, cur, i = 0;
    for(; i<n; ++i){
        key = seq.substr(i, length);
        cur = freq.get(key);
        cur === undefined ? freq.set(key, new RefNum(1)) : ++cur.num;
    }
    return freq;
}

function sort(seq, length){
    var f = frequency(seq, length), keys = Array.from(f.keys()),
        n = seq.length-length+1, res = '';
    keys.sort((a, b)=>f.get(b)-f.get(a));
    for (var key of keys) res += key+' '+(f.get(key)*100/n).toFixed(3)+'\n';
    res += '\n';
    return res;
}

function find(seq, s){
    var f = frequency(seq, s.length);
    return (f.get(s) || 0)+"\t"+s+'\n';
}

function master() {
    var workers = [], reading = false, seq = '', maxBuf = 2048;

    for (var i=0; i<4;++i) workers.push(cluster.fork({workerId: i}));

    var flush = function(close) {
        for (var w of workers) w.send(seq);
        if (close) for (var w of workers) w.send('eof');
        seq = '';
    };
    var lineHandler = function(line){
        if (reading) {
            if (line[0] !== '>' &&
                    (seq += line.toUpperCase()).length >= maxBuf) flush();
        } else reading = line.substr(0, 6) === '>THREE';
    };
    rd.createInterface(process.stdin, process.stdout)
        .on('line', lineHandler).on('close', function() { flush(true); });

    var jobs = workers.length, results = [];
    var messageHandler = function(i){
        return function(message){
            results[i] = message;
            if (!(--jobs)) {
                process.stdout.write(results.join(''));
                process.exit(0);
            }
        };
    };
    for (var i=0; i<workers.length; ++i)
        workers[i].on('message', messageHandler(i));
}

function worker(){
    var seq = '';
    process.on('message', function(message){
        if (message === 'eof') {
            var res = '';
            switch (process.env.workerId) {
                case '0':
                    res += sort(seq, 1);
                    res += sort(seq, 2);
                    res += find(seq, "GGT");
                    break;
                case '1':
                    res += find(seq, "GGTA");
                    res += find(seq, "GGTATT");
                    break;
                case '2':
                    res += find(seq, "GGTATTTTAATT");
                    break;
                case '3':
                    res += find(seq, "GGTATTTTAATTTATAGT");
                    break;
            }
            process.send(res);
            process.exit();
        } else seq += message;
    });
}

cluster.isMaster ? master() : worker();
    

notes, command-line, and program output

NOTES:
64-bit Ubuntu quad core
v7.9.0


Fri, 14 Apr 2017 23:37:42 GMT

MAKE:
cp -L knucleotide.node knucleotide.js
0.21s to complete and log all make actions

COMMAND LINE:
/usr/local/src/node-v7.9.0-linux-x64/bin/node knucleotide.js 0 < knucleotide-input25000000.txt

PROGRAM OUTPUT:
A 30.295
T 30.151
C 19.800
G 19.754

AA 9.177
TA 9.132
AT 9.131
TT 9.091
CA 6.002
AC 6.001
AG 5.987
GA 5.984
CT 5.971
TC 5.971
GT 5.957
TG 5.956
CC 3.917
GC 3.911
CG 3.909
GG 3.902

1471758	GGT
446535	GGTA
47336	GGTATT
893	GGTATTTTAATT
893	GGTATTTTAATTTATAGT