Revision 412c3dbfb208525e7fa8d2885a956e8295721507, 2.6 kB (checked in by Theo Schlossnagle <>, 4 years ago)

Actual native docbook.

  • Property mode set to 100644
1 <?xml version="1.0" encoding="UTF-8"?>
2 <?xml-model href="" schematypens=""?>
3 <?xml-model href="" type="application/xml" schematypens=""?>
4 <preface xmlns="" xml:id="introduction" version="5">
5   <title>Introduction</title>
6   <para>
7     Reconnoiter is a monitoring and trend analysis system designed to cope
8     with large architectures (thousands of machines and hundreds of thousands
9     of metrics).
10   </para>
11   <para>
12     Heavy focus is placed on decoupling the various components of the system
13     to allow for disjoint evolution of each component as issues arise or new
14     requirements are identified.  Resource monitoring, metric aggregation,
15     metric analysis and visualization are all cleanly separated.
16   </para>
17   <para>
18     The monitor, <command>noitd</command>, is written in C and designed to
19     support highly concurrent and rapid checks with an expected capability
20     of monitoring 100,000 services per minute (6 million checks per hour.) 
21     While it is hard to make writing checks "easy" in this high-performance
22     environment, efforts have been made to ensure that custom check scripting
23     does not require the expertise of writing highly-concurrent, event-driven
24     C code.  Instead, glue is provided via scripting languages such as Lua
25     that attempt to handle aspects of this high-concurrency environment
26     transparently.  As with any high-performance system, you can easily
27     introduce non-performant code and jeopardize performance system-wide.
28   </para>
29   <para>
30     The aggregator, <command>stratcond</command>, is also written in C and
31     responsible for the simple task of securely gathering data from all of the
32     distributed <command>noitd</command> instances and transforming them into
33     the data storage facility (currently PostgreSQL).
34   </para>
35   <para>
36     The data storage facility (PostgreSQL) holds all information about
37     individual checks, their statuses and the individual metrics associated
38     with them.  Automatic processes are in place that summarize the
39     numeric metrics into windowed averages for expedient graphing at
40     a variety of time window resolutions (hour, day, month, year, etc.).
41   </para>
42   <para>
43     The visualization interface (reconnoiter) draws from the data store
44     to visualize collected metrics and assist with monitoring, trending and
45     other visual analyses.  The visualization system is written in PHP.
46   </para>
47 </preface>
48 <!--
49 vim:ts=2:sw=2:et:
50 -->
Note: See TracBrowser for help on using the browser.