blob: b6b5942a53bd65443327a69cdb69e15000d7441c [file] [log] [blame]
Ian Maxon858061a2022-05-12 19:11:28 -07001<!DOCTYPE html>
2<!--
3 | Generated by Apache Maven Doxia Site Renderer 1.8.1 from src/site/markdown/ncservice.md at 2022-05-12
4 | Rendered using Apache Maven Fluido Skin 1.7
5-->
6<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
7 <head>
8 <meta charset="UTF-8" />
9 <meta name="viewport" content="width=device-width, initial-scale=1.0" />
10 <meta name="Date-Revision-yyyymmdd" content="20220512" />
11 <meta http-equiv="Content-Language" content="en" />
12 <title>AsterixDB &#x2013; Installation using NCService</title>
13 <link rel="stylesheet" href="./css/apache-maven-fluido-1.7.min.css" />
14 <link rel="stylesheet" href="./css/site.css" />
15 <link rel="stylesheet" href="./css/print.css" media="print" />
16 <script type="text/javascript" src="./js/apache-maven-fluido-1.7.min.js"></script>
17
18 </head>
19 <body class="topBarDisabled">
20 <div class="container-fluid">
21 <div id="banner">
22 <div class="pull-left"><a href="./" id="bannerLeft"><img src="images/asterixlogo.png" alt="AsterixDB"/></a></div>
23 <div class="pull-right"></div>
24 <div class="clear"><hr/></div>
25 </div>
26
27 <div id="breadcrumbs">
28 <ul class="breadcrumb">
29 <li id="publishDate">Last Published: 2022-05-12</li>
30 <li id="projectVersion" class="pull-right">Version: 0.9.8</li>
31 <li class="pull-right"><a href="index.html" title="Documentation Home">Documentation Home</a></li>
32 </ul>
33 </div>
34 <div class="row-fluid">
35 <div id="leftColumn" class="span2">
36 <div class="well sidebar-nav">
37 <ul class="nav nav-list">
38 <li class="nav-header">Get Started - Installation</li>
39 <li class="active"><a href="#"><span class="none"></span>Option 1: using NCService</a></li>
40 <li><a href="ansible.html" title="Option 2: using Ansible"><span class="none"></span>Option 2: using Ansible</a></li>
41 <li><a href="aws.html" title="Option 3: using Amazon Web Services"><span class="none"></span>Option 3: using Amazon Web Services</a></li>
42 <li class="nav-header">AsterixDB Primer</li>
43 <li><a href="sqlpp/primer-sqlpp.html" title="Using SQL++"><span class="none"></span>Using SQL++</a></li>
44 <li class="nav-header">Data Model</li>
45 <li><a href="datamodel.html" title="The Asterix Data Model"><span class="none"></span>The Asterix Data Model</a></li>
46 <li class="nav-header">Queries</li>
47 <li><a href="sqlpp/manual.html" title="The SQL++ Query Language"><span class="none"></span>The SQL++ Query Language</a></li>
48 <li><a href="SQLPP.html" title="Raw SQL++ Grammar"><span class="none"></span>Raw SQL++ Grammar</a></li>
49 <li><a href="sqlpp/builtins.html" title="Builtin Functions"><span class="none"></span>Builtin Functions</a></li>
50 <li class="nav-header">API/SDK</li>
51 <li><a href="api.html" title="HTTP API"><span class="none"></span>HTTP API</a></li>
52 <li><a href="csv.html" title="CSV Output"><span class="none"></span>CSV Output</a></li>
53 <li class="nav-header">Advanced Features</li>
54 <li><a href="aql/externaldata.html" title="Accessing External Data"><span class="none"></span>Accessing External Data</a></li>
55 <li><a href="feeds.html" title="Data Ingestion with Feeds"><span class="none"></span>Data Ingestion with Feeds</a></li>
56 <li><a href="udf.html" title="User Defined Functions"><span class="none"></span>User Defined Functions</a></li>
57 <li><a href="sqlpp/filters.html" title="Filter-Based LSM Index Acceleration"><span class="none"></span>Filter-Based LSM Index Acceleration</a></li>
58 <li><a href="sqlpp/fulltext.html" title="Support of Full-text Queries"><span class="none"></span>Support of Full-text Queries</a></li>
59 <li><a href="sqlpp/similarity.html" title="Support of Similarity Queries"><span class="none"></span>Support of Similarity Queries</a></li>
60 <li><a href="geo/quickstart.html" title="GIS Support Overview"><span class="none"></span>GIS Support Overview</a></li>
61 <li><a href="geo/functions.html" title="GIS Functions"><span class="none"></span>GIS Functions</a></li>
62 <li><a href="interval_join.html" title="Support of Interval Joins"><span class="none"></span>Support of Interval Joins</a></li>
63 <li><a href="spatial_join.html" title="Support of Spatial Joins"><span class="none"></span>Support of Spatial Joins</a></li>
64 <li><a href="sqlpp/arrayindex.html" title="Support of Array Indexes"><span class="none"></span>Support of Array Indexes</a></li>
65 <li class="nav-header">Deprecated</li>
66 <li><a href="aql/primer.html" title="AsterixDB Primer: Using AQL"><span class="none"></span>AsterixDB Primer: Using AQL</a></li>
67 <li><a href="aql/manual.html" title="Queries: The Asterix Query Language (AQL)"><span class="none"></span>Queries: The Asterix Query Language (AQL)</a></li>
68 <li><a href="aql/builtins.html" title="Queries: Builtin Functions (AQL)"><span class="none"></span>Queries: Builtin Functions (AQL)</a></li>
69</ul>
70 <hr />
71 <div id="poweredBy">
72 <div class="clear"></div>
73 <div class="clear"></div>
74 <div class="clear"></div>
75 <div class="clear"></div>
76<a href="./" title="AsterixDB" class="builtBy"><img class="builtBy" alt="AsterixDB" src="images/asterixlogo.png" /></a>
77 </div>
78 </div>
79 </div>
80 <div id="bodyColumn" class="span10" >
81<!--
82 ! Licensed to the Apache Software Foundation (ASF) under one
83 ! or more contributor license agreements. See the NOTICE file
84 ! distributed with this work for additional information
85 ! regarding copyright ownership. The ASF licenses this file
86 ! to you under the Apache License, Version 2.0 (the
87 ! "License"); you may not use this file except in compliance
88 ! with the License. You may obtain a copy of the License at
89 !
90 ! http://www.apache.org/licenses/LICENSE-2.0
91 !
92 ! Unless required by applicable law or agreed to in writing,
93 ! software distributed under the License is distributed on an
94 ! "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
95 ! KIND, either express or implied. See the License for the
96 ! specific language governing permissions and limitations
97 ! under the License.
98 !-->
99<h1>Installation using NCService</h1>
100<div class="section">
101<h2><a name="Table_of_Contents"></a><a name="toc" id="toc">Table of Contents</a></h2>
102<ul>
103
104<li><a href="#quickstart">Quick Start</a></li>
105<li><a href="#Small_cluster">Starting a small single-machine cluster using the NCService</a></li>
106<li><a href="#Multi_machine">Deploying AsterixDB via NCService in a multi-machine setup</a></li>
107<li><a href="#Parameters">Available Configuration Parameters</a></li>
108</ul>
109<h1><a name="quickstart" id="quickstart">Quick Start</a></h1>
110<p>The fastest way to get set up with a single-machine sample instance of AsterixDB is to use the included sample helper scripts. To do so, in the extracted <tt>asterix-server</tt> directory, navigate to <tt>opt/local/bin/</tt></p>
111
112<div>
113<div>
114<pre class="source">user@localhost:~/
115$cd asterix-server/
116user@localhost:~/asterix-server
117$cd opt/local/bin
118</pre></div></div>
119
120<p>This folder should contain 4 scripts, two pairs of <tt>.sh</tt> and <tt>.bat</tt> files respectively. <tt>start-sample-cluster.sh</tt> will simply start a basic sample cluster using the configuration files located in <tt>opt/local/conf/</tt>.</p>
121
122<div>
123<div>
124<pre class="source">user@localhost:~/a/o/l/bin
125$./start-sample-cluster.sh
126CLUSTERDIR=/home/user/asterix-server/opt/local
127INSTALLDIR=/home/user/asterix-server
128LOGSDIR=/home/user/asterix-server/samples/opt/logs
129
130INFO: Starting sample cluster...
131INFO: Waiting up to 30 seconds for cluster 127.0.0.1:19002 to be available.
132INFO: Cluster started and is ACTIVE.
133user@localhost:~/a/o/l/bin
134$
135</pre></div></div>
136
137<p>Now, there should be a running AsterixDB cluster on the machine. To go to the Web Interface, visit <a class="externalLink" href="http://localhost:19001">http://localhost:19001</a></p>
138
139<div class="source">
140
141<div class="source"><pre class="prettyprint linenums">
142<img src="images/asterixdb_interface.png" alt="The AsterixDB Web Interface" />
143<i>Fig. 1</i>: The AsterixDB Web Interface
144</pre></div>
145</div>
146
147<h1><a name="Small_cluster" id="Small_cluster">Starting a small single-machine cluster using NCService</a></h1>
148<p>The above cluster was started using a script, but below is a description in detail of how precisely this was achieved. The config files here are analagous to the ones within <tt>samples/local/conf</tt>.</p>
149<p>When running a cluster using the <tt>NCService</tt> there are 3 different kinds of processes involved:</p>
150<ul>
151
152<li><tt>NCDriver</tt>, also known as the Node Controller or NC for short. This is the process that does the actual work of queries and data management within the AsterixDB cluster</li>
153<li><tt>NCService</tt> configures and starts the <tt>NCDriver</tt> process. It is a simple daemon whose only purpose is to wait for the <tt>CCDriver</tt> process to call upon it to initiate cluster bootup.</li>
154<li><tt>CCDriver</tt>, which is the Cluster Controller process, also known as the CC. This process manages the distribution of tasks to all NCs, as well as providing the parameters of each NC to the NCService upon cluster startup. It also hosts the Web interface and query compiler and optimizer.</li>
155</ul>
156<p>The cluster startup follows a particular sequence, which is as follows:</p>
157<ol style="list-style-type: decimal">
158
159<li>Each host on which an NC is desired and is mentioned in the configuration, the <tt>NCService</tt> daemon is started first. It will listen and wait for the CC to contact it.</li>
160<li>The one host on which the CC is to be placed is started with an appropriate configuration file.</li>
161<li>The CC contacts all <tt>NCService</tt> daemons and the <tt>NCService</tt> subsequently starts and <tt>NCDriver</tt> process with the configration supplied by the <tt>CC</tt></li>
162<li>Each <tt>NCDriver</tt> then contacts the CC to register itself as started</li>
163</ol>
164<p>This process is briefly illustrated in the diagram below:</p>
165
166<div class="source">
167
168<div class="source"><pre class="prettyprint linenums">
169<img src="images/ncservice.png" alt="The AsterixDB Web Interface" />
170<i>Fig. 2</i>: NCService startup sequence
171</pre></div>
172</div>
173
174<p>To start a small cluster consisting of 2 NodeControllers (<tt>red</tt> and <tt>blue</tt>) and 1 ClusterController (<tt>cc</tt>) on a single machine only 2 configuration files are required. The first one is</p>
175<p><tt>blue.conf</tt>:</p>
176
177<div>
178<div>
179<pre class="source">[ncservice]
180port=9091
181</pre></div></div>
182
183<p>It is a configuration file for the second <tt>NCService</tt>. This contains only the port that the <tt>NCService</tt> of the second NodeControllers listens to as it is non-standard. The first <tt>NCService</tt> does not need a configuration file, as it only uses default parameters. In a distributed environment with 1 NodeController per machine, no <tt>NCService</tt> needs a configuration file.</p>
184<p>The second configuration file is</p>
185<p><tt>cc.conf</tt>:</p>
186
187<div>
188<div>
189<pre class="source">[nc/red]
190txn.log.dir=/tmp/asterix/red/txnlog
191core.dump.dir=/tmp/asterix/red/coredump
192iodevices=/tmp/asterix/red
193
194[nc/blue]
195port=9091
196txn.log.dir=/tmp/asterix/blue/txnlog
197core.dump.dir=/tmp/asterix/blue/coredump
198iodevices=/tmp/asterix/blue
199
200[nc]
201app.class=org.apache.asterix.hyracks.bootstrap.NCApplicationEntryPoint
202address=127.0.0.1
203command=asterixnc
204
205[cc]
206address = 127.0.0.1
207console.listen.port = 12345
208</pre></div></div>
209
210<p>This is the configuration file for the cluster and it contains information that each <tt>NCService</tt> will use when starting the corresponding <tt>NCDriver</tt> as well as information for the <tt>CCDriver</tt>.</p>
211<p>To start the cluster simply use the following steps</p>
212<ol style="list-style-type: decimal">
213
214<li>
215
216<p>Change directory into the asterix-server binary folder</p>
217
218<div>
219<div>
220<pre class="source">user@localhost:~/
221$cd asterix-server/
222user@localhost:~/asterix-server
223$cd samples/local/bin
224</pre></div></div>
225</li>
226<li>
227
228<p>Start the 2 <tt>NCServices</tt> for <tt>red</tt> and <tt>blue</tt>.</p>
229
230<div>
231<div>
232<pre class="source">user@localhost:~/asterix-server
233$bin/asterixncservice -config-file blue.conf &gt; blue-service.log 2&gt;&amp;1 &amp;
234user@localhost:~/asterix-server
235$bin/asterixncservice &gt;red-service.log 2&gt;&amp;1 &amp;
236</pre></div></div>
237</li>
238<li>
239
240<p>Start the <tt>CCDriver</tt>.</p>
241
242<div>
243<div>
244<pre class="source">user@localhost:~/asterix-server
245$bin/asterixcc -config-file cc.conf &gt; cc.log 2&gt;&amp;1 &amp;
246</pre></div></div>
247</li>
248</ol>
249<p>The <tt>CCDriver</tt> will connect to the <tt>NCServices</tt> and thus initiate the configuration and the start of the <tt>NCDrivers</tt>. After running these scripts, <tt>jps</tt> should show a result similar to this:</p>
250
251<div>
252<div>
253<pre class="source">user@localhost:~/asterix-server
254$jps
25513184 NCService
25613200 NCDriver
25713185 NCService
25813186 CCDriver
25913533 Jps
26013198 NCDriver
261</pre></div></div>
262
263<p>The logs for the <tt>NCDrivers</tt> will be in <tt>$BASEDIR/logs</tt>.</p>
264<p>To stop the cluster again simply run</p>
265
266<div>
267<div>
268<pre class="source">$ kill `jps | egrep '(CDriver|NCService)' | awk '{print $1}'`
269</pre></div></div>
270
271<p>to kill all processes.</p>
272<h1><a name="Multi_machine" id="Multi_machine">Deploying AsterixDB via NCService in a multi-machine setup</a></h1>
273<p>Deploying on multiple machines only differs in the configuration file and where each process is actually resident. Take for example a deployment on 3 machines, <tt>cacofonix-1</tt>,<tt>cacofonix-2</tt>,and <tt>cacofonix-3</tt>. <tt>cacofonix-1</tt> will be the CC, and <tt>cacofonix-2</tt> and <tt>cacofonix-3</tt> will be the two NCs, respectively. The configuration would be as follows:</p>
274<p><tt>cc.conf</tt>:</p>
275
276<div>
277<div>
278<pre class="source">[nc/red]
279txn.log.dir=/lv_scratch/asterix/red/txnlog
280core.dump.dir=/lv_scratch/asterix/red/coredump
281iodevices=/lv_scratch/asterix/red
282address=cacofonix-2
283
284[nc/blue]
285txn.log.dir=/lv_scratch/asterix/blue/txnlog
286core.dump.dir=/lv_scratch/asterix/blue/coredump
287iodevices=/lv_scratch/asterix/blue
288address=cacofonix-3
289
290[nc]
291app.class=org.apache.asterix.hyracks.bootstrap.NCApplicationEntryPoint
292storagedir=storage
293command=asterixnc
294
295[cc]
296address = cacofonix-1
297</pre></div></div>
298
299<p>To deploy, first the <tt>asterix-server</tt> binary must be present on each machine. Any method to transfer the archive to each machine will work, but here <tt>scp</tt> will be used for simplicity&#x2019;s sake.</p>
300
301<div>
302<div>
303<pre class="source">user@localhost:~
304$for f in {1,2,3}; scp asterix-server.zip cacofonix-$f:~/; end
305</pre></div></div>
306
307<p>Then unzip the binary on each machine. First, start the <tt>NCService</tt> processes on each of the slave machines. Any way of getting a shell on the machine is fine, be it physical or via <tt>ssh</tt>.</p>
308
309<div>
310<div>
311<pre class="source">user@cacofonix-2 12:41:42 ~/asterix-server/
312$ bin/asterixncservice &gt; red-service.log 2&gt;&amp;1 &amp;
313
314
315user@cacofonix-3 12:41:42 ~/asterix-server/
316$ bin/asterixncservice &gt; blue-service.log 2&gt;&amp;1 &amp;
317</pre></div></div>
318
319<p>Now that each <tt>NCService</tt> is waiting, the CC can be started.</p>
320
321<div>
322<div>
323<pre class="source">user@cacofonix-1 12:41:42 ~/asterix-server/
324$ bin/asterixcc -config-file cc.conf &gt; cc.log 2&gt;&amp;1 &amp;
325</pre></div></div>
326
327<p>The cluster should now be started and the Web UI available on the CC host at the default port.</p>
328<h1><a name="Parameters" id="Parameters">Available Configuration Parameters</a></h1>
329<p>The following parameters are for the master process, under the &#x201c;[cc]&#x201d; section.</p>
330<table border="0" class="table table-striped">
331<thead>
332
333<tr class="a">
334<th> Section </th>
335<th> Parameter </th>
336<th> Meaning </th>
337<th> Default </th></tr>
338</thead><tbody>
339
340<tr class="b">
341<td> cc </td>
342<td> active.port </td>
343<td> The listen port of the active server </td>
344<td> 19003 </td></tr>
345<tr class="a">
346<td> cc </td>
347<td> address </td>
348<td> Default bind address for all services on this cluster controller </td>
349<td> 127.0.0.1 </td></tr>
350<tr class="b">
351<td> cc </td>
352<td> api.port </td>
353<td> The listen port of the API server </td>
354<td> 19002 </td></tr>
355<tr class="a">
356<td> cc </td>
357<td> app.class </td>
358<td> Application CC main class </td>
359<td> org.apache.asterix.hyracks.bootstrap.CCApplication </td></tr>
360<tr class="b">
361<td> cc </td>
362<td> client.listen.address </td>
363<td> Sets the IP Address to listen for connections from clients </td>
364<td> same as address </td></tr>
365<tr class="a">
366<td> cc </td>
367<td> client.listen.port </td>
368<td> Sets the port to listen for connections from clients </td>
369<td> 1098 </td></tr>
370<tr class="b">
371<td> cc </td>
372<td> cluster.listen.address </td>
373<td> Sets the IP Address to listen for connections from NCs </td>
374<td> same as address </td></tr>
375<tr class="a">
376<td> cc </td>
377<td> cluster.listen.port </td>
378<td> Sets the port to listen for connections from node controllers </td>
379<td> 1099 </td></tr>
380<tr class="b">
381<td> cc </td>
382<td> cluster.public.address </td>
383<td> Address that NCs should use to contact this CC </td>
384<td> same as cluster.listen.address </td></tr>
385<tr class="a">
386<td> cc </td>
387<td> cluster.public.port </td>
388<td> Port that NCs should use to contact this CC </td>
389<td> same as cluster.listen.port </td></tr>
390<tr class="b">
391<td> cc </td>
392<td> cluster.topology </td>
393<td> Sets the XML file that defines the cluster topology </td>
394<td> &lt;undefined&gt; </td></tr>
395<tr class="a">
396<td> cc </td>
397<td> console.listen.address </td>
398<td> Sets the listen address for the Cluster Controller </td>
399<td> same as address </td></tr>
400<tr class="b">
401<td> cc </td>
402<td> console.listen.port </td>
403<td> Sets the http port for the Cluster Controller) </td>
404<td> 16001 </td></tr>
405<tr class="a">
406<td> cc </td>
407<td> cores.multiplier </td>
408<td> The factor to multiply by the number of cores to determine maximum query concurrent execution level </td>
409<td> 3 </td></tr>
410<tr class="b">
411<td> cc </td>
412<td> heartbeat.max.misses </td>
413<td> Sets the maximum number of missed heartbeats before a node is marked as dead </td>
414<td> 5 </td></tr>
415<tr class="a">
416<td> cc </td>
417<td> heartbeat.period </td>
418<td> Sets the time duration between two heartbeats from each node controller in milliseconds </td>
419<td> 10000 </td></tr>
420<tr class="b">
421<td> cc </td>
422<td> job.history.size </td>
423<td> Limits the number of historical jobs remembered by the system to the specified value </td>
424<td> 10 </td></tr>
425<tr class="a">
426<td> cc </td>
427<td> job.manager.class </td>
428<td> Specify the implementation class name for the job manager </td>
429<td> org.apache.hyracks.control.cc.job.JobManager </td></tr>
430<tr class="b">
431<td> cc </td>
432<td> job.queue.capacity </td>
433<td> The maximum number of jobs to queue before rejecting new jobs </td>
434<td> 4096 </td></tr>
435<tr class="a">
436<td> cc </td>
437<td> job.queue.class </td>
438<td> Specify the implementation class name for the job queue </td>
439<td> org.apache.hyracks.control.cc.scheduler.FIFOJobQueue </td></tr>
440<tr class="b">
441<td> cc </td>
442<td> profile.dump.period </td>
443<td> Sets the time duration between two profile dumps from each node controller in milliseconds; 0 to disable </td>
444<td> 0 </td></tr>
445<tr class="a">
446<td> cc </td>
447<td> result.sweep.threshold </td>
448<td> The duration within which an instance of the result cleanup should be invoked in milliseconds </td>
449<td> 60000 </td></tr>
450<tr class="b">
451<td> cc </td>
452<td> result.ttl </td>
453<td> Limits the amount of time results for asynchronous jobs should be retained by the system in milliseconds </td>
454<td> 86400000 </td></tr>
455<tr class="a">
456<td> cc </td>
457<td> root.dir </td>
458<td> Sets the root folder used for file operations </td>
459<td> ${java.io.tmpdir}/asterixdb/ClusterControllerService </td></tr>
460<tr class="b">
461<td> cc </td>
462<td> web.port </td>
463<td> The listen port of the legacy query interface </td>
464<td> 19001 </td></tr>
465<tr class="a">
466<td> cc </td>
467<td> web.queryinterface.port </td>
468<td> The listen port of the query web interface </td>
469<td> 19006 </td></tr>
470</tbody>
471</table>
472<p>The following parameters for slave processes, under &#x201c;[nc]&#x201d; sections.</p>
473<table border="0" class="table table-striped">
474<thead>
475
476<tr class="a">
477<th> Section </th>
478<th> Parameter </th>
479<th> Meaning </th>
480<th> Default </th></tr>
481</thead><tbody>
482
483<tr class="b">
484<td> nc </td>
485<td> address </td>
486<td> Default IP Address to bind listeners on this NC. All services will bind on this address unless a service-specific listen address is supplied. </td>
487<td> 127.0.0.1 </td></tr>
488<tr class="a">
489<td> nc </td>
490<td> app.class </td>
491<td> Application NC Main Class </td>
492<td> org.apache.asterix.hyracks.bootstrap.NCApplication </td></tr>
493<tr class="b">
494<td> nc </td>
495<td> cluster.address </td>
496<td> Cluster Controller address (required unless specified in config file) </td>
497<td> &lt;undefined&gt; </td></tr>
498<tr class="a">
499<td> nc </td>
500<td> cluster.connect.retries </td>
501<td> Number of attempts to contact CC before giving up </td>
502<td> 5 </td></tr>
503<tr class="b">
504<td> nc </td>
505<td> cluster.listen.address </td>
506<td> IP Address to bind cluster listener on this NC </td>
507<td> same as address </td></tr>
508<tr class="a">
509<td> nc </td>
510<td> cluster.listen.port </td>
511<td> IP port to bind cluster listener </td>
512<td> 0 </td></tr>
513<tr class="b">
514<td> nc </td>
515<td> cluster.port </td>
516<td> Cluster Controller port </td>
517<td> 1099 </td></tr>
518<tr class="a">
519<td> nc </td>
520<td> cluster.public.address </td>
521<td> Public IP Address to announce cluster listener </td>
522<td> same as public.address </td></tr>
523<tr class="b">
524<td> nc </td>
525<td> cluster.public.port </td>
526<td> Public IP port to announce cluster listener </td>
527<td> same as cluster.listen.port </td></tr>
528<tr class="a">
529<td> nc </td>
530<td> command </td>
531<td> Command NCService should invoke to start the NCDriver </td>
532<td> hyracksnc </td></tr>
533<tr class="b">
534<td> nc </td>
535<td> core.dump.dir </td>
536<td> The directory where node core dumps should be written </td>
537<td> ${java.io.tmpdir}/asterixdb/coredump </td></tr>
538<tr class="a">
539<td> nc </td>
540<td> data.listen.address </td>
541<td> IP Address to bind data listener </td>
542<td> same as address </td></tr>
543<tr class="b">
544<td> nc </td>
545<td> data.listen.port </td>
546<td> IP port to bind data listener </td>
547<td> 0 </td></tr>
548<tr class="a">
549<td> nc </td>
550<td> data.public.address </td>
551<td> Public IP Address to announce data listener </td>
552<td> same as public.address </td></tr>
553<tr class="b">
554<td> nc </td>
555<td> data.public.port </td>
556<td> Public IP port to announce data listener </td>
557<td> same as data.listen.port </td></tr>
558<tr class="a">
559<td> nc </td>
560<td> iodevices </td>
561<td> Comma separated list of IO Device mount points </td>
562<td> ${java.io.tmpdir}/asterixdb/iodevice </td></tr>
563<tr class="b">
564<td> nc </td>
565<td> jvm.args </td>
566<td> JVM args to pass to the NCDriver </td>
567<td> &lt;undefined&gt; </td></tr>
568<tr class="a">
569<td> nc </td>
570<td> messaging.listen.address </td>
571<td> IP Address to bind messaging listener </td>
572<td> same as address </td></tr>
573<tr class="b">
574<td> nc </td>
575<td> messaging.listen.port </td>
576<td> IP port to bind messaging listener </td>
577<td> 0 </td></tr>
578<tr class="a">
579<td> nc </td>
580<td> messaging.public.address </td>
581<td> Public IP Address to announce messaging listener </td>
582<td> same as public.address </td></tr>
583<tr class="b">
584<td> nc </td>
585<td> messaging.public.port </td>
586<td> Public IP port to announce messaging listener </td>
587<td> same as messaging.listen.port </td></tr>
588<tr class="a">
589<td> nc </td>
590<td> ncservice.address </td>
591<td> Address the CC should use to contact the NCService associated with this NC </td>
592<td> same as public.address </td></tr>
593<tr class="b">
594<td> nc </td>
595<td> ncservice.pid </td>
596<td> PID of the NCService which launched this NCDriver </td>
597<td> -1 </td></tr>
598<tr class="a">
599<td> nc </td>
600<td> ncservice.port </td>
601<td> Port the CC should use to contact the NCService associated with this NC </td>
602<td> 9090 </td></tr>
603<tr class="b">
604<td> nc </td>
605<td> net.buffer.count </td>
606<td> Number of network buffers per input/output channel </td>
607<td> 1 </td></tr>
608<tr class="a">
609<td> nc </td>
610<td> net.thread.count </td>
611<td> Number of threads to use for Network I/O </td>
612<td> 1 </td></tr>
613<tr class="b">
614<td> nc </td>
615<td> public.address </td>
616<td> Default public address that other processes should use to contact this NC. All services will advertise this address unless a service-specific public address is supplied. </td>
617<td> same as address </td></tr>
618<tr class="a">
619<td> nc </td>
620<td> result.listen.address </td>
621<td> IP Address to bind dataset result distribution listener </td>
622<td> same as address </td></tr>
623<tr class="b">
624<td> nc </td>
625<td> result.listen.port </td>
626<td> IP port to bind dataset result distribution listener </td>
627<td> 0 </td></tr>
628<tr class="a">
629<td> nc </td>
630<td> result.manager.memory </td>
631<td> Memory usable for result caching at this Node Controller in bytes </td>
632<td> -1 (-1 B) </td></tr>
633<tr class="b">
634<td> nc </td>
635<td> result.public.address </td>
636<td> Public IP Address to announce dataset result distribution listener </td>
637<td> same as public.address </td></tr>
638<tr class="a">
639<td> nc </td>
640<td> result.public.port </td>
641<td> Public IP port to announce dataset result distribution listener </td>
642<td> same as result.listen.port </td></tr>
643<tr class="b">
644<td> nc </td>
645<td> result.sweep.threshold </td>
646<td> The duration within which an instance of the result cleanup should be invoked in milliseconds </td>
647<td> 60000 </td></tr>
648<tr class="a">
649<td> nc </td>
650<td> result.ttl </td>
651<td> Limits the amount of time results for asynchronous jobs should be retained by the system in milliseconds </td>
652<td> 86400000 </td></tr>
653<tr class="b">
654<td> nc </td>
655<td> storage.buffercache.maxopenfiles </td>
656<td> The maximum number of open files in the buffer cache </td>
657<td> 2147483647 </td></tr>
658<tr class="a">
659<td> nc </td>
660<td> storage.buffercache.pagesize </td>
661<td> The page size in bytes for pages in the buffer cache </td>
662<td> 131072 (128 kB) </td></tr>
663<tr class="b">
664<td> nc </td>
665<td> storage.buffercache.size </td>
666<td> The size of memory allocated to the disk buffer cache. The value should be a multiple of the buffer cache page size. </td>
667<td> 1/4 of the JVM allocated memory </td></tr>
668<tr class="a">
669<td> nc </td>
670<td> storage.lsm.bloomfilter.falsepositiverate </td>
671<td> The maximum acceptable false positive rate for bloom filters associated with LSM indexes </td>
672<td> 0.01 </td></tr>
673<tr class="b">
674<td> nc </td>
675<td> storage.memorycomponent.globalbudget </td>
676<td> The size of memory allocated to the memory components. The value should be a multiple of the memory component page size </td>
677<td> 1/4 of the JVM allocated memory </td></tr>
678<tr class="a">
679<td> nc </td>
680<td> storage.memorycomponent.numcomponents </td>
681<td> The number of memory components to be used per lsm index </td>
682<td> 2 </td></tr>
683<tr class="b">
684<td> nc </td>
685<td> storage.memorycomponent.pagesize </td>
686<td> The page size in bytes for pages allocated to memory components </td>
687<td> 131072 (128 kB) </td></tr>
688<tr class="a">
689<td> nc </td>
690<td> storage.metadata.memorycomponent.numpages </td>
691<td> The number of pages to allocate for a metadata memory component </td>
692<td> 8 </td></tr>
693<tr class="b">
694<td> nc </td>
695<td> txn.log.dir </td>
696<td> The directory where transaction logs should be stored </td>
697<td> ${java.io.tmpdir}/asterixdb/txn-log </td></tr>
698</tbody>
699</table>
700<p>The following parameters are configured under the &#x201c;[common]&#x201d; section.</p>
701<table border="0" class="table table-striped">
702<thead>
703
704<tr class="a">
705<th> Section </th>
706<th> Parameter </th>
707<th> Meaning </th>
708<th> Default </th></tr>
709</thead><tbody>
710
711<tr class="b">
712<td> common </td>
713<td> active.memory.global.budget </td>
714<td> The memory budget (in bytes) for the active runtime </td>
715<td> 67108864 (64 MB) </td></tr>
716<tr class="a">
717<td> common </td>
718<td> compiler.framesize </td>
719<td> The page size (in bytes) for computation </td>
720<td> 32768 (32 kB) </td></tr>
721<tr class="b">
722<td> common </td>
723<td> compiler.groupmemory </td>
724<td> The memory budget (in bytes) for a group by operator instance in a partition </td>
725<td> 33554432 (32 MB) </td></tr>
726<tr class="a">
727<td> common </td>
728<td> compiler.joinmemory </td>
729<td> The memory budget (in bytes) for a join operator instance in a partition </td>
730<td> 33554432 (32 MB) </td></tr>
731<tr class="b">
732<td> common </td>
733<td> compiler.parallelism </td>
734<td> The degree of parallelism for query execution. Zero means to use the storage parallelism as the query execution parallelism, while other integer values dictate the number of query execution parallel partitions. The system will fall back to use the number of all available CPU cores in the cluster as the degree of parallelism if the number set by a user is too large or too small </td>
735<td> 0 </td></tr>
736<tr class="a">
737<td> common </td>
738<td> compiler.sortmemory </td>
739<td> The memory budget (in bytes) for a sort operator instance in a partition </td>
740<td> 33554432 (32 MB) </td></tr>
741<tr class="b">
742<td> common </td>
743<td> compiler.sort.parallel </td>
744<td> Enable full parallel sort for queries </td>
745<td> true </td></tr>
746<tr class="a">
747<td> common </td>
748<td> compiler.sort.samples </td>
749<td> The number of samples taken from each partition to guide the sort operation when full parallel sort is enabled </td>
750<td> 100 </td></tr>
751<tr class="b">
752<td> common </td>
753<td> compiler.textsearchmemory </td>
754<td> The memory budget (in bytes) for an inverted-index-search operator instance in a partition </td>
755<td> 33554432 (32 MB) </td></tr>
756<tr class="a">
757<td> common </td>
758<td> compiler.windowmemory </td>
759<td> The memory budget (in bytes) for a window operator instance in a partition </td>
760<td> 33554432 (32 MB) </td></tr>
761<tr class="b">
762<td> common </td>
763<td> log.level </td>
764<td> The logging level for master and slave processes </td>
765<td> WARNING </td></tr>
766<tr class="a">
767<td> common </td>
768<td> max.wait.active.cluster </td>
769<td> The max pending time (in seconds) for cluster startup. After the threshold, if the cluster still is not up and running, it is considered unavailable </td>
770<td> 60 </td></tr>
771<tr class="b">
772<td> common </td>
773<td> messaging.frame.count </td>
774<td> Number of reusable frames for NC to NC messaging </td>
775<td> 512 </td></tr>
776<tr class="a">
777<td> common </td>
778<td> messaging.frame.size </td>
779<td> The frame size to be used for NC to NC messaging </td>
780<td> 4096 (4 kB) </td></tr>
781<tr class="b">
782<td> common </td>
783<td> metadata.callback.port </td>
784<td> IP port to bind metadata callback listener (0 = random port) </td>
785<td> 0 </td></tr>
786<tr class="a">
787<td> common </td>
788<td> metadata.listen.port </td>
789<td> IP port to bind metadata listener (0 = random port) </td>
790<td> 0 </td></tr>
791<tr class="b">
792<td> common </td>
793<td> metadata.node </td>
794<td> the node which should serve as the metadata node </td>
795<td> &lt;undefined&gt; </td></tr>
796<tr class="a">
797<td> common </td>
798<td> metadata.registration.timeout.secs </td>
799<td> how long in seconds to wait for the metadata node to register with the CC </td>
800<td> 60 </td></tr>
801<tr class="b">
802<td> common </td>
803<td> replication.log.batchsize </td>
804<td> The size in bytes to replicate in each batch </td>
805<td> 4096 (4 kB) </td></tr>
806<tr class="a">
807<td> common </td>
808<td> replication.log.buffer.numpages </td>
809<td> The number of log buffer pages </td>
810<td> 8 </td></tr>
811<tr class="b">
812<td> common </td>
813<td> replication.log.buffer.pagesize </td>
814<td> The size in bytes of each log buffer page </td>
815<td> 131072 (128 kB) </td></tr>
816<tr class="a">
817<td> common </td>
818<td> replication.max.remote.recovery.attempts </td>
819<td> The maximum number of times to attempt to recover from a replica on failure before giving up </td>
820<td> 5 </td></tr>
821<tr class="b">
822<td> common </td>
823<td> replication.timeout </td>
824<td> The time in seconds to timeout when trying to contact a replica, before assuming it is dead </td>
825<td> 15 </td></tr>
826<tr class="a">
827<td> common </td>
828<td> storage.max.active.writable.datasets </td>
829<td> The maximum number of datasets that can be concurrently modified </td>
830<td> 8 </td></tr>
831<tr class="b">
832<td> common </td>
833<td> txn.commitprofiler.enabled </td>
834<td> Enable output of commit profiler logs </td>
835<td> false </td></tr>
836<tr class="a">
837<td> common </td>
838<td> txn.commitprofiler.reportinterval </td>
839<td> Interval (in seconds) to report commit profiler logs </td>
840<td> 5 </td></tr>
841<tr class="b">
842<td> common </td>
843<td> txn.job.recovery.memorysize </td>
844<td> The memory budget for each job job (in bytes) used for recovery </td>
845<td> 67108864 (64 MB) </td></tr>
846<tr class="a">
847<td> common </td>
848<td> txn.lock.escalationthreshold </td>
849<td> The maximum number of entity locks to obtain before upgrading to a dataset lock </td>
850<td> 1000 </td></tr>
851<tr class="b">
852<td> common </td>
853<td> txn.lock.shrinktimer </td>
854<td> The time (in milliseconds) where under utilization of resources will trigger a shrink phase </td>
855<td> 5000 </td></tr>
856<tr class="a">
857<td> common </td>
858<td> txn.lock.timeout.sweepthreshold </td>
859<td> Interval (in milliseconds) for checking lock timeout </td>
860<td> 10000 </td></tr>
861<tr class="b">
862<td> common </td>
863<td> txn.lock.timeout.waitthreshold </td>
864<td> Time out (in milliseconds) of waiting for a lock </td>
865<td> 60000 </td></tr>
866<tr class="a">
867<td> common </td>
868<td> txn.log.buffer.numpages </td>
869<td> The number of pages in the transaction log tail </td>
870<td> 8 </td></tr>
871<tr class="b">
872<td> common </td>
873<td> txn.log.buffer.pagesize </td>
874<td> The page size (in bytes) for transaction log buffer </td>
875<td> 4194304 (4MB) </td></tr>
876<tr class="a">
877<td> common </td>
878<td> txn.log.checkpoint.history </td>
879<td> The number of checkpoints to keep in the transaction log </td>
880<td> 0 </td></tr>
881<tr class="b">
882<td> common </td>
883<td> txn.log.checkpoint.lsnthreshold </td>
884<td> The checkpoint threshold (in terms of LSNs (log sequence numbers) that have been written to the transaction log, i.e., the length of the transaction log) for transaction logs </td>
885<td> 67108864 (64 MB) </td></tr>
886<tr class="a">
887<td> common </td>
888<td> txn.log.checkpoint.pollfrequency </td>
889<td> The frequency (in seconds) the checkpoint thread should check to see if a checkpoint should be written </td>
890<td> 120 </td></tr>
891<tr class="b">
892<td> common </td>
893<td> txn.log.partitionsize </td>
894<td> The maximum size (in bytes) of each transaction log file </td>
895<td> 268435456 (256 MB) </td></tr>
896</tbody>
897</table>
898<p>For the optional NCService process configuration file, the following parameters, under &#x201c;[ncservice]&#x201d; section.</p>
899<table border="0" class="table table-striped">
900<thead>
901
902<tr class="a">
903<th> Parameter </th>
904<th> Meaning </th>
905<th> Default </th></tr>
906</thead><tbody>
907
908<tr class="b">
909<td> address </td>
910<td> The address the NCService listens on for commands from the CC </td>
911<td> (all addresses) </td></tr>
912<tr class="a">
913<td> port </td>
914<td> The port for the NCService listens on for commands from the CC </td>
915<td> 9090 </td></tr>
916<tr class="b">
917<td> logdir </td>
918<td> Directory where NCService logs should be written (&#x2018;-&#x2019; indicates that output should go to stdout) </td>
919<td> ${app.home}/logs (${user.home} if &#x2018;app.home&#x2019; not present in NCService Java system properties. </td></tr>
920</tbody>
921</table></div>
922 </div>
923 </div>
924 </div>
925 <hr/>
926 <footer>
927 <div class="container-fluid">
928 <div class="row-fluid">
929<div class="row-fluid">Apache AsterixDB, AsterixDB, Apache, the Apache
930 feather logo, and the Apache AsterixDB project logo are either
931 registered trademarks or trademarks of The Apache Software
932 Foundation in the United States and other countries.
933 All other marks mentioned may be trademarks or registered
934 trademarks of their respective owners.
935 </div>
936 </div>
937 </div>
938 </footer>
939 </body>
940</html>