Infobright installation and use of the source code

Brief introduction:

Infobright is a database based on patented technology, a MySQL-based open source data Warehouse software that can be used as a storage engine for MySQL, and select queries are no different from normal MySQL.
first, several major advantages:
1, high compression ratio, the average compression ratio of up to 10:1. (tested, our 15 million 697M log data compression ratio is 6:1, compressed data size is only 114M)
2, column storage, even if the amount of data is very large, query speed is fast. (tested, a compound query in Infobright to 30 seconds, in MySQL to more than a point)
3, does not need to build index, avoids maintains the index and the index as the data expands the question. Each column of data block compressed storage, each block has a knowledge grid node record block statistics, in place of the index, speed up the search.
4. Single server can read and write 30T data efficiently. Extensible, this means that for the same query, when the amount of data is 10T, it should not consume more time than 1T data volume, basically is an order of magnitude.

Note: Actually Infobright takes a record a lot slower than MySQL, but it takes 100W records faster than MySQL. Therefore, it is more useful for logging, or for summarizing large amounts of data.

second, infobright restrictions
1, does not support the Data Update: Community Edition Infobright can only use "LOAD data INFILE" way to import data, does not support insert, UPDATE, DELETE
2, does not support high concurrency: can only support more than 10-18 concurrent queries

Install dependent software:

1, boost   

wgetHttp://ncu.dl.sourceforge.net/project/boost/boost/1.49.0/boost_1_49_0.tar.gz
#tar zxvf boost_1_49_0.tar.gz
#./bootstrap.sh--prefix=/usr/local/boost_1_49_0
#./bjam Install
#export boost_root=/usr/local/boost_1_49_0

2,ncurses

Yum-y Install ncurses Ncurses-devel

Infobright installation:

Https://www.infobright.org/downloads/ice/infobright-4.0.7-0-src-ice.tar.gz

cd infobright-4.0.7make  prefix=/usr/local/infobright  edition=community  Releasemake prefix=/usr/local/infobright   edition=community install-releasemkdir  /usr/local/infobright/conf  /usr/local/infobright/data /usr/local/infobright/logschown  -r mysql.mysql /usr/local/infobright/data /usr/local/infobright/logscp src/build/ pkgmt/my-ib.cnf /usr/local/infobright/conf/my-ib.cnf/usr/local/infobright/bin/mysql_install_db -- Basedir=/usr/local/mysql --datadir=/usr/local/infobright/data --user=mysql 

Modify MY-IB.CNF

Basedir =/usr/local/infobrightdatadir =/usr/local/infobright/datalog-error =/usr/local/infobright/logs/bh.err

Launch IB instance

Cd/usr/local/infobrightbin/mysqld_safe--defaults-file=conf/my-ib.cnf--user=mysql >/dev/null 2>&1 &

Initialize the password of the IB instance

/usr/local/infobright/bin/mysqladmin-u root Password "123456"

Because of the ice community version used, you can only import data using IB loader (in fact, it is only supported by manually importing data into CSV files).
Note: CREATE TABLE t () Engine=brighthouse xxx;
When you create a table, the engine of the table uses "BrightHouse" to use the features of the IB warehouse (because infobright-4.0.7 has a MySQL storage engine such as MyISAM, memory, etc., if not specified, it is possible to use other engines)
For example, suppose the CSV data file is Data.csv

Bin/mysql-u root-p123456load Data infile "/root/data.csv" into table t_data fields terminated by ', ' enclosed by ' "Esc aped by ' \ ' lines terminated by ' \ n ';

Once the data has been imported successfully, you can perform data mining operations such as SUM, AVG, group BY, and so on in IB.

Infobright installation and use of the source code