MySQL Storage Engines - page 6
What is a Storage Engine?
The FEDERATED storage engine (added in MySQL 5.03) enables you to access data from remote MySQL database (other databases may be supported in the future) as if it were a local database. In effect, the MySQL server acts as a proxy to the remote server, using the MySQL client access library to connect to the remote host, execute queries and then reformat the data into the localized format.
Table 5. FEDERATED Summary
|Data limitations||Limited by remote database|
The ARCHIVE storage engine supports only the INSERT and SELECT statements, but does support most of the MySQL field types. Information stored in an ARCHIVE storage engine table is compressed and cannot be modified and so ARCHIVE tables are perfect for storing log data (which you don't want to be able to change) or information that is no longer in active use (for example, old invoicing or sales data).
While the information is stored very efficient, care should be taken when accessing data stored in the ARCHIVE tables. Because the information is compressed, selects have to read the entire table, and that also means decompressing the information. This can obviously increase the time taken to perform complex searches and retrievals. If you are performing a large number of queries on the information in these tables it may be easier to temporarily copy your data to another, uncompressed, data type such as MyISAM.
Table 6. ARCHIVE Summary
|Data limitations||Data can only be inserted (no updates)|
The CSV storage engine stores data not in a binary format, but in the form a CSV (Command Separated Values) file. Because of this, there are limitations to the data stored. It is not an efficient method for storing large volumes of data, or larger data types like BLOB, although such types are supported. There is also no indexing. However, because the data is stored in the CSV format it is exceedingly portable; these CSV files generated can easily be imported into many different software packages, including Excel, OpenOffice and database systems like Access or FileMaker.
In general, the CSV engine is impractical as a general database engine. It is, however, probably the most effective and easiest method for data exchange. What makes it so convenient is that we can use SELECT and INSERT statements to create the database, which in turn means that we can easily produce CSV files based on queries of other data.
With some careful work, the CSV storage engine can also be used as an effective way of getting information into MySQL. Here, you can create the tables first, shutdown the MySQL server, copy over CSV files that you have exported from Excel, Access or another database, and you can then import the data and copy it over to MyISAM or InnoDB tables.
Table 7. CSV Summary
|Index limitations||Indexing is not supported|
Strange though it may seem, the BLACKHOLE engine does not actually store any data. Although you can create tables and indexes, all SQL statements that would add or update information to the database are executed without actually writing any data. The database structure is retained, however, and you can create any indexes on the (non-existent) information that you want.
Although this seems like a futile exercise, it does allow you to test out database structures and play with table definitions without actually creating any data. Even more useful, however, is that SQL statements on BLACKHOLE databases are written to the binary log, and therefore are replicated to slave databases.
You can use this functionality to update one or more slaves directly without writing any local data. There are a number of potential uses for this functionality. One such use I have employed in past is to write log data to a BLACKHOLE table, which is then echoed to two slaves. Because the write is instantaneous (there are no local disk files or indexes to update), I can maintain a high logging rate, and rely on the binary logging and slave replication to distribute the data. An extension of this, as suggested in the MySQL manual, is to use filtering to control the distribution of records to the slaves.
Table 8. BLACKHOLE Summary
|Data limitations||No data is stored, but statements are written to the binary log (and therefore distributed to slave databases)|
Solid state disks (SSDs) made a splash in consumer technology, and now the technology has its eyes on the enterprise storage market. Download this eBook to see what SSDs can do for your infrastructure and review the pros and cons of this potentially game-changing storage technology.
- 1Linux Top 3: Ubuntu 14.04, Debian Gives Squeeze More Life and Red Hat Goes Atomic
- 2Linux Top 3: CoreOS, Oracle Enterprise Linux 7 and Ubuntu 14.10
- 3Linux Top 3: Debian Dumps SPARC, Ubuntu Takes Over Linux 3.13 and the Core Infrastructure Initiative
- 4Linux Top 3: Fedora, Ubuntu and Gluster Lose Community Leaders
- 5Red Hat Enterprise Linux 7 Finally Hits the Big Time