2 Overview

The management of data in telecommunications systems has many aspects of which some, but not all, are addressed by traditional Database Management Systems (DBMSs). In particular, the high level of fault tolerance required in many nonstop systems, combined with requirements on the DBMS to run in the same address space as the applications, have led us to implement a new DBMS, called Mnesia.

Mnesia is implemented in, and tightly coupled to Erlang. It provides the functionality that is necessary for the implementation of fault-tolerant telecommunications systems.

Mnesia is a multiuser distributed DBMS specifically designed for industrial-grade telecommunications applications written in Erlang, which is also the intended target language. Mnesia tries to address all the data management issues required for typical telecommunications systems and has a number of features not normally found in traditional DBMSs.

Telecommunications applications need a mix of a broad range of features generally not provided by traditional DBMSs. Mnesia is designed to meet requirements such as:

• Fast real-time key-value lookup
• Complex non-real-time queries (mainly for operation and maintenance tasks)
• Distributed data (due to the distributed nature of the applications)
• High fault tolerance
• Dynamic reconfiguration
• Complex objects

Mnesia addresses the typical data management issues required for telecommunications applications which sets it apart from most other DBMSs. It combines many concepts found in traditional DBMSs, such as transactions and queries, with concepts found in data management systems for telecommunications applications such as:

• Fast real-time operations
• Configurable replication for fault tolerance
• Dynamic reconfiguration without service disruption

Mnesia is also unique due to its tight coupling to Erlang. It almost turns Erlang into a database programming language, which yields many benefits. The foremost is that the impedance mismatch between the data format used by the DBMS and the data format used by the programming language, which is used to manipulate the data, completely disappears.

2.1 The Mnesia Database Management System

Features

Mnesia has the following features that combine to produce a fault-tolerant distributed database management system (DBMS) written in Erlang:

• Database schema can be dynamically reconfigured at runtime.
• Tables can be declared to have properties such as location, replication, and persistence.
• Tables can be moved or replicated to several nodes to improve fault tolerance. Other nodes in the system can still access the tables to read, write, and delete records.
• Table locations are transparent to the programmer. Programs address table names and the system itself keeps track of table locations.
• Transactions can be distributed and multiple operations can be executed within a single transaction.
• Multiple transactions can run concurrently and their execution is fully synchronized by Mnesia, ensuring that no two processes manipulate the same data simultaneously.
• Transactions can be assigned the property of being executed on all nodes in the system, or on none.
• Transactions can be bypassed using dirty operations, which reduce overheads and run fast.

All of the above features are described in detail in the coming sections.

Query List Comprehension

Query List Comprehension (QLC) can be used with Mnesia to produce specialized functions that enhance its operational ability. QLC has its own documentation as part of the OTP documentation set. The main QLC advantages when used with Mnesia are:

• QLC can optimize the query compiler for Mnesia, essentially making the system more efficient.
• QLC can be used as a database programming language for Mnesia. It includes a notation called list comprehensions which can be used to execute complex database queries over a set of tables.

For more information about QLC, please see the qlc manual page in STDLIB.

When to Use Mnesia

Mnesia is a great fit for applications that:

• Need to replicate data.
• Perform complex data queries.
• Need to use atomic transactions to safely update several records simultaneously.
• Require soft real-time characteristics.

Mnesia is not as appropriate for applications that:

• Process plain text or binary data files.
• Merely need a lookup dictionary that can be stored on disc. Such applications may use the standard library module dets, which is a disc-based version of the ets module. For more information about dets, please see the dets manual page in STDLIB.
• Need disc logging facilities. Such applications may use the module disk_log. For more information about disk_log, please see the disk_log manual page in Kernel.
• Require hard real-time characteristics.