Mimer SQL has products both for 32-bit Linux systems and for 64-bit Linux systems.
From Mimer SQL version 10, Linux systems based the kernel version 3 or later are supported – as far as operating system revisions are backwards compatible. Please, see the section Supported Architecturesbelow for details on supported and unsupported platforms.
For features in Mimer SQL, see the article Mimer SQL overview as a starting point.
Mimer SQL provides small footprint, scalable and robust relational database solutions that conform to international ISO SQL standards. It is very well suited for high performance mission critical systems as well as for mobile and embedded appliances. In addition, Mimer SQL is equipped with an extensive multilingual support using collations.
A Mimer SQL Engine distribution contains several components, as described below.
A default license for testing and development is included, it allows up to ten concurrent users. For license upgrades, please contact Mimer SQL sales.
The Mimer SQL database server is a single, multi-threaded process with SMP scalability.
Clients using TCP/IP can access the server. For clients running on the same platform, a shared-memory based communication method is used.
A Module SQL preprocessor is included. It enables separation of SQL code and a host application written in C, FORTRAN, COBOL or Pascal into different source files, simplifying modularity and reuse of SQL code.
A JDBC driver is included in the distribution. The driver is a type 4 driver, which means that it is written entirely in Java. This provides the driver with full portability so that it can be copied or downloaded to any Java enabled platform. The driver uses TCP/IP to access a Mimer SQL server (version 8.2 or later) on any platform. For details on the JDBC drivers, please refer to the Mimer JDBC Driver Guide found at our documentation page.
The Mimer ODBC driver is a client library that enables applications to access Mimer SQL database servers running on any platform. The driver complies with the ODBC 3.52 specification. In the Linux ODBC environment it may vary how SQLWCHAR is defined and used, i.e. if it refers to a two or four byte type. Mimer SQL on Linux supports a four byte SQLWCHAR in the libmimodbc shared library.
On Linux you can choose to use a third party ODBC Driver Manager that enables applications to dynamically load drivers for different database products. But, you can also choose to link your applications directly to the Mimer ODBC driver, without using any Driver Manager. In the latter case we suggest usage of the provided ODBC header files, introduced by including the mimcli.h file.
For a proper Unicode behavior, please note that internationalized programs must include the locale.h header file and call the setlocale() operating system function to initiate a specific language operation. This can be done by calling setlocale() as follows:
Generally when executing an ODBC application, a platform specific Mimer ODBC driver has to be installed on the client side. This driver can then access Mimer database servers on any platform. There is no need to install any special software on the server side in order to use ODBC.
Mimer SQL C API
Mimer SQL C API is a native C library suitable for tool integration and application development in environments where API standardization is not a requirement. The following characteristics describe the API:
- Platform independence
- Small footprint
- Tight fit with the Mimer SQL application/database communication model.
The so called “MimerAPI” is provided in the libmimerapi.so shared library and the mimerapi.h header file provides prototypes and other handy defines. See Database API article for MimerAPI and the Mimer SQL Programming Manual (found in the Mimer SQL Documentation Set at the Documentation page.
The Mimer SQL product for Linux includes the Mimer SQL manuals in Adobe PDF format. Online documentation can be found at the Mimer SQL documentation page.
The main utility used when administering and managing a Mimer SQL database is the DbVisualizer Java based tool, which is provided within the delivered product package.
Mimer SQL also includes command-line utilities for creating databases, load and unload of data, backups, interactive SQL statement execution, etc.
In addition, the Mimer SQL distribution contains an example database, examples of database programs and SQL statements.
This product runs on processors built on the Instruction Set Architecture (ISA) called IA-32, or x86-32. Several suppliers are providing processors that supports IA-32, for example, Intel (Pentium, Celeron, Xeon, …) and AMD (Athlon, Duron, Sempron, …). Mimer SQL for Linux supports all of these.
In addition, an extension to IA-32 is available called AMD64, or x86-64. This architecture is backwards compatible with IA-32. Accordingly, Mimer SQL for Linux can be used on processors based on AMD64. For example, the Athlon-64 and Opteron processors from AMD supports AMD64.
Intel has also developed processors supporting the AMD64 architecture. They have named their variant Intel 64. This architecture is compatible with AMD64, and consequently, it is supported by Mimer SQL for Linux. Processors here are those with names like Core 2 Duo, Core 2 Quad and Core i7 series.
AMD64 (x86-64) and Intel 64 shall not be mixed up with the IA-64 architecture (Itanium), developed by Intel. Itanium is incompatible with IA-32, and thus, this version of Mimer SQL for Linux does not work on Itanium.
An AMD64 processor started in 64-bit mode can run 64-bit and 32-bit applications at the same time, but if started in 32-bit mode it cannot run 64-bit applications until it has been restarted. Please note that Mimer SQL is currently not available in 64-bit mode which implies that 64-bit programs cannot be linked with Mimer SQL, and that a larger Mimer SQL cache than 4Gb cannot be used. But still, the IA-32 based Mimer SQL for Linux runs perfectly on the platform, assuming 32-bit versions of the dynamic link libraries are installed.