首先要感谢Dave先生的贡献才有了redis_fdw。Dave贡献了两个FDW模块，mysql_fdw和redis_fdw分别用于PostgreSQL9.1连接到MySQL和Redis。
接下来通过redis_fdw来测试一下PostgreSQL 9.1是怎么连接到Redis的。
首先到
http://redis.io下载一个redis2.4.1的稳定版源码。
安装参考README
加上密码校验，并放置后台运行，用作测试其他参数不修改了。
启动redis
确认已启动
[root@digoal redis-2.4.1]# tar -zxvf redis-2.4.1.tar.gz
[root@digoal redis-2.4.1]# cd redis-2.4.1
[root@digoal redis-2.4.1]# make
[root@digoal redis-2.4.1]# vi redis.conf加上密码校验，并放置后台运行，用作测试其他参数不修改了。
daemonize yes
requirepass DIGOAL启动redis
[root@digoal redis-2.4.1]# src/redis-server ./redis.conf 确认已启动
[root@digoal redis-2.4.1]# netstat -anp|grep 6379
tcp 0 0 0.0.0.0:6379 0.0.0.0:* LISTEN 26647/redis-server
接下来在另一台PostgreSQL数据库服务器上配置redis_fdw.
首先是下载redis_fdw
http://www.pgxn.org/user/pgsnake
解压到postgresql源码包的 postgresql-9.1.0/contrib/ 目录
然后把redis-2.4.1/deps/hiredis 目录拷贝到 postgresql-9.1.0/contrib/redis_fdw-1.0.0
接下来编译hiredis
确认安装OK之后。编译redis_fdw
OK，至此就算安装好了。
cd hiredis
make
make install确认安装OK之后。编译redis_fdw
[root@digoal redis_fdw-1.0.0]# . /home/postgres/.bash_profile
(导入两个环境变量 PATH, LD_LIBRARY_PATH)
PATH和LD_LIBRARY_PATH如下 :
export PGHOME=/opt/pgsql91
export LD_LIBRARY_PATH=$PGHOME/lib:/lib64:/usr/lib64:/usr/local/lib64:/lib:/usr/lib:/usr/local/lib
export PATH=$PGHOME/bin:$PATH:.
[root@digoal redis_fdw-1.0.0]# USE_PGXS=1 make clean
[root@digoal redis_fdw-1.0.0]# USE_PGXS=1 make
[root@digoal redis_fdw-1.0.0]# USE_PGXS=1 make install
-- 以下步骤可能可以省略, 如果没有才需要拷贝.
[root@digoal redis_fdw-1.0.0]# cp redis_fdw.so /opt/pgsql/lib/
[root@digoal redis_fdw-1.0.0]# cp redis_fdw--1.0.sql /opt/pgsql/share/extension/
[root@digoal redis_fdw-1.0.0]# cp redis_fdw.control /opt/pgsql/share/extension/OK，至此就算安装好了。
接下来体验一下。
先到PostgreSQL数据库里面创建redis_fdw这个extension .
然后到redis服务端插入一些数据。
注意事项请参考redis_fdw README.
psql -h 127.0.0.1 digoal postgres
psql (9.1.0)
Type "help" for help.
digoal=# create extension redis_fdw;然后到redis服务端插入一些数据。
vi ins.sh
set visits:digoal1:totals 1
set visits:digoal2:totals 2
set visits:digoal3:totals 3
set visits:digoal4:totals 4
set visits:digoal5:totals 5
set visits:digoal6:totals 6
set visits:digoal7:totals 7
set visits:digoal8:totals 8
set visits:digoal9:totals 9
set visits:digoal10:totals 10
set visits:digoal11:totals 11
set visits:digoal12:totals 12
set visits:digoal13:totals 13
set visits:digoal14:totals 14
set visits:digoal15:totals 15
set visits:digoal16:totals 16
set visits:digoal17:totals 17
set visits:digoal18:totals 18
set visits:digoal19:totals 19
set visits:digoal20:totals 20
set visits:digoal21:totals 21
set visits:digoal22:totals 22
set visits:digoal23:totals 23
set visits:digoal24:totals 24
set visits:digoal25:totals 25
set visits:digoal26:totals 26
set visits:digoal27:totals 27
set visits:digoal28:totals 28
set visits:digoal29:totals 29
set visits:digoal30:totals 30
set visits:digoal31:totals 31
set visits:digoal32:totals 32
set visits:digoal33:totals 33
set visits:digoal34:totals 34
set visits:digoal35:totals 35
set visits:digoal36:totals 36
set visits:digoal37:totals 37
set visits:digoal38:totals 38
set visits:digoal39:totals 39
set visits:digoal40:totals 40
set visits:digoal41:totals 41
set visits:digoal42:totals 42
set visits:digoal43:totals 43
set visits:digoal44:totals 44
set visits:digoal45:totals 45
set visits:digoal46:totals 46
set visits:digoal47:totals 47
set visits:digoal48:totals 48
set visits:digoal49:totals 49
set visits:digoal50:totals 50
set visits:digoal51:totals 51
set visits:digoal52:totals 52
set visits:digoal53:totals 53
set visits:digoal54:totals 54
set visits:digoal55:totals 55
set visits:digoal56:totals 56
set visits:digoal57:totals 57
set visits:digoal58:totals 58
set visits:digoal59:totals 59
set visits:digoal60:totals 60
set visits:digoal61:totals 61
set visits:digoal62:totals 62
set visits:digoal63:totals 63
set visits:digoal64:totals 64
set visits:digoal65:totals 65
set visits:digoal66:totals 66
set visits:digoal67:totals 67
set visits:digoal68:totals 68
set visits:digoal69:totals 69
set visits:digoal70:totals 70
set visits:digoal71:totals 71
set visits:digoal72:totals 72
set visits:digoal73:totals 73
set visits:digoal74:totals 74
set visits:digoal75:totals 75
set visits:digoal76:totals 76
set visits:digoal77:totals 77
set visits:digoal78:totals 78
set visits:digoal79:totals 79
set visits:digoal80:totals 80
set visits:digoal81:totals 81
set visits:digoal82:totals 82
set visits:digoal83:totals 83
set visits:digoal84:totals 84
set visits:digoal85:totals 85
set visits:digoal86:totals 86
set visits:digoal87:totals 87
set visits:digoal88:totals 88
set visits:digoal89:totals 89
set visits:digoal90:totals 90
set visits:digoal91:totals 91
set visits:digoal92:totals 92
set visits:digoal93:totals 93
set visits:digoal94:totals 94
set visits:digoal95:totals 95
set visits:digoal96:totals 96
set visits:digoal97:totals 97
set visits:digoal98:totals 98
set visits:digoal99:totals 99
set visits:digoal100:totals 100
导入到库0和库1
[root@digoal src]# cat ./ins.sh|./redis-cli -h 127.0.0.1 -p 6379 -a DIGOAL -n 0
[root@digoal src]# cat ./ins.sh|./redis-cli -h 127.0.0.1 -p 6379 -a DIGOAL -n 1
接下来在PostgreSQL数据库创建连接REDIS的外部表：
digoal=# CREATE SERVER redis_server1
FOREIGN DATA WRAPPER redis_fdw
OPTIONS (address '172.xxx.xxx.xxx', port '6379');
digoal=# CREATE USER MAPPING FOR digoal
SERVER redis_server1
OPTIONS (password 'DIGOAL');
CREATE USER MAPPING
digoal=# grant usage on FOREIGN server redis_server1 to digoal;
GRANT
digoal=> CREATE FOREIGN TABLE redis_db0 (key text, value text)
SERVER redis_server1
OPTIONS (database '0');
CREATE FOREIGN TABLE
digoal=> CREATE FOREIGN TABLE redis_db1 (key text, value text)
SERVER redis_server1
OPTIONS (database '1');
CREATE FOREIGN TABLE
digoal=> select * from redis_db0 limit 5;
key | value
------------------------+-------
visits:digoal13:totals | 13
visits:digoal23:totals | 23
visits:digoal33:totals | 33
visits:digoal43:totals | 43
visits:digoal17:totals | 17
(5 rows)
digoal=> select * from redis_db1 limit 5;
key | value
------------------------+-------
visits:digoal13:totals | 13
visits:digoal23:totals | 23
visits:digoal33:totals | 33
visits:digoal43:totals | 43
visits:digoal17:totals | 17
(5 rows)
digoal=> select count(*) from redis_db0;
count
-------
100
(1 row)
digoal=> select count(*) from redis_db1;
count
-------
100
(1 row)注意事项请参考redis_fdw README.
【参考】
【其他FDW】
1. hive
2. PostgreSQL Foreign Table - pgsql_fdw
3. PostgreSQL Foreign Table - oracle_fdw 1
4. PostgreSQL Foreign Table - oracle_fdw 2
5. PostgreSQL Foreign Table - oracle_fdw 3
6. PostgreSQL Foreign Table - file_fdw
7. PostgreSQL Foreign Table - redis_fdw
8. PostgreSQL Foreign Table - mysql_fdw 1
9. PostgreSQL Foreign Table - mysql_fdw 2
