dataminer

Dataminer

Dataminer is a data mining, fault-tolerant distributed worker queue built for creating job workers simple and quickly. Common routines for processing queues, streaming data into other queues or distributing queues, flush routines into other data sources are all apart of dataminer's core tasks. Dataminer is backed by redis and mongodb.

$ npm install dataminer

Features

---

• fault-tolerant queues
• worker monitoring
• queue events and progress
• worker specific logging
• powered by Redis & MongoDB
• restful json api
• fault tolerant streams
• exponential and linear backoff

Creating Workers

---

First to create any standard queue worker that processes items off of a redis queue, use dataminer.createQueue.

dataminer.createQueue(queueName, n, options)

Create a new queue worker to process on queueName. n is the number of calls that will be done at any given time for each job (n = 2 will process 2 jobs at a time for a single process). n defaults to 1. options are extended properties for the worker that include:

• name: A friendly name for a given queue worker (defaults to queueName provided from parameters.
• redis: Redis related options for processing the queue and optionally reporting on the status of the worker.
  • host: defaults to localhost.
  • port: defaults to 6379.
  • options: redis options to the redis client.
  • auth: optional auth parameters to pass to the redis client.
• report: defaults to true to report on the status of the worker once every reportInterval milliseconds.
• reportIntervalMs: defaults to 1000
• progress: defaults to false to update progress with a traceable job id.

var dataminer = require('dataminer'),
    request = require('request');

var downloader = dataminer.createQueue('q-urls', { progress: true });
downloader.process(function (job, data, done) {

    var contentLength = 0;
    var req = request(data.url);
    req.pipe(fs.createWriteStream(data.path));
    req.on('response', function (response) {
        contentLength = response.headers['content-length'];
    });
    req.on('data', function (chunk) {
        job.progress(chunk.length, contentLength, false, 'active');
    });
    req.on('error', function (err) {
        job.progress(0, contentLength, false, 'failed');
        done(err);
    });
    req.on('end', function () {
        job.progress(0, contentLength, true, 'done');
        done();
    }):

});

• dataminer.createQueue('q-urls') or with progress: false will result in a callback function that does not include the job but will instead only include the data and done callback. This is useful for higher-throughput processing that requires no progress tracking such as parsing tweets.

downloader.process(function (data, done) {
});

dataminer.register(worker, id, options)

Registers a worker to a specific id so that when the process dies the status of the worker will be in a stopped state along with any other remnant items such as the processed count per second history.

dataminer.register(worker, 'dkdksfa323id');

dataminer.unregister(id, options)

Unregisters a given worker id so that all previous remnant artifacts such as the worker process count per second history, the status..etc are immediately removed from redis.

dataminer.unregister('oldunusedworkerid');

dataminer.app.listen(options, port, fn)

Starts the http express web application to expose the json api. Use this to configure additional parameters for auth or additional routes and configurations. The default api is located at /api/v1/workers e.g.

dataminer.app.listen()

dataminer.download(path, url, n, job, done)

Runs an async download requests from url and saves to the path. n represents the number of async requests to run if the url supports the accepts-range header. The accepts-range header allows the requested file to be broken up into byte chunks and downloaded appropriately. job can be used in addition if this method is called from within a worker so that progress can be automatically tracked. Finally, done is the callback handler for when the download is completed.

dataminer.download('test.pdf', 'http://kindle.s3.amazonaws.com/Kindle_Touch_Users_Guide.pdf', 2)

JSON API

---

The JSON API included with the project allows you to efficiently monitor your workers and optionally load items into the queue, restart workers and other native options to the api. The endpoints are described below:

• /api/v1/workers: Returns a list of active/inactive workers running across all instances and monitored services.
• /api/v1/jobs: Returns a list of actively running jobs across all instances and monitored services.

TODO:

---

• Add streaming to queue support ala dataminer.createStream
  • Add Twitter Sample Stream Example (use request.pipe)
• Add flush routine support to existing queue workers
  • queueWorker.flush(flushIntervalMs, function () { })
• Add cluster awareness support
  • Monitor each other's running services to either run workers in a "single-failover" scenario or load-balanced.
  • Automatically start up failover workers
  • Make a group of running workers configurable (nameable)
    • i.e. Parsers vs Streams
  • Spawn up a new set of workers based on group name
  • Spawn up a worker group to be in failover vs load-balanced mode
• Use process.hrtime() in cases of timing intervals
• Add tracer logging support
• Add winston logging support

LICENSE:

---

(The MIT License)

Copyright (c) 2012 nyxtom@gmail.com

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the 'Software'), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED 'AS IS', WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.