比如有如下表需要以crt_time為標準按月分割槽

create table gf_card (
  id            bigint not null,
  crt_time        datetime not null,
  name            varchar(20) not null，
  CONSTRAINT PK_card_id PRIMARY KEY (id asc)
)

(1).一般會將表按每個月的資料放入不同的檔案組中，先給每個月建一個檔案組

--建立檔案組
alter database test add filegroup fileGroup01
alter database test add filegroup fileGroup02
alter database test add filegroup fileGroup03
alter database test add filegroup fileGroup04
alter database test add filegroup fileGroup05
alter database test add filegroup fileGroup06
alter database test add filegroup fileGroup07
alter database test add filegroup fileGroup08
alter database test add filegroup fileGroup09
alter database test add filegroup fileGroup10
alter database test add filegroup fileGroup11
alter database test add filegroup fileGroup12
 

如果要刪除檔案組可使用：

--刪除檔案組
alter database test remove filegroup  fileGroup01

(2).一個檔案組可以包含多個檔案，為了提高效能檔案一般放在多個不同的物理盤上，建立了檔案組後需要指定這些檔案組中包含哪些檔案

---建立資料檔案到檔案組
alter database test add file (name='test01',filename=N'D:\tmp\data\test01.ndf',size=5Mb,filegrowth=5mb) to filegroup fileGroup01
alter database test add file (name='test02',filename=N'D:\tmp\data\test02.ndf',size=5Mb,filegrowth=5mb) to filegroup fileGroup02
alter database test add file (name='test03',filename=N'D:\tmp\data\test03.ndf',size=5Mb,filegrowth=5mb) to filegroup fileGroup03
alter database test add file (name='test04',filename=N'D:\tmp\data\test04.ndf',size=5Mb,filegrowth=5mb) to filegroup fileGroup04
alter database test add file (name='test05',filename=N'D:\tmp\data\test05.ndf',size=5Mb,filegrowth=5mb) to filegroup fileGroup05
alter database test add file (name='test06',filename=N'D:\tmp\data\test06.ndf',size=5Mb,filegrowth=5mb) to filegroup fileGroup06
alter database test add file (name='test07',filename=N'D:\tmp\data\test07.ndf',size=5Mb,filegrowth=5mb) to filegroup fileGroup07
alter database test add file (name='test08',filename=N'D:\tmp\data\test08.ndf',size=5Mb,filegrowth=5mb) to filegroup fileGroup08
alter database test add file (name='test09',filename=N'D:\tmp\data\test09.ndf',size=5Mb,filegrowth=5mb) to filegroup fileGroup09
alter database test add file (name='test10',filename=N'D:\tmp\data\test10.ndf',size=5Mb,filegrowth=5mb) to filegroup fileGroup10
alter database test add file (name='test11',filename=N'D:\tmp\data\test11.ndf',size=5Mb,filegrowth=5mb) to filegroup fileGroup11
alter database test add file (name='test12',filename=N'D:\tmp\data\test12.ndf',size=5Mb,filegrowth=5mb) to filegroup fileGroup12
 

如果要刪除資料檔案到檔案組可使用：

---刪除資料檔案到檔案組
alter database test remove file  test01

(3). 分割槽函式是為了劃分分割槽邊界，range left表示{，value]，(value，}，range right表示{，value)，[value，}

--建立分割槽函式
create partition function [PF_card] (datetime) as range right for values
('2017-01-01','2017-02-01','2017-03-01','2017-04-01','2017-05-01','2017-06-01',
'2017-07-01','2017-08-01','2017-09-01','2017-10-01','2017-11-01','2017-12-01')
 

如果要刪除分割槽函式可使用：

--刪除分割槽函式
drop partition function [PF_card]

(4).有了分割槽函式後，需要指定不同分割槽的資料放到哪個檔案組中

--建立分割槽方案
CREATE PARTITION SCHEME [PS_card] AS PARTITION [PF_card]
TO ([PRIMARY], [fileGroup01], [fileGroup02], [fileGroup03], [fileGroup04], [fileGroup05], 
[fileGroup06], [fileGroup07], [fileGroup08], [fileGroup09], [fileGroup10], [fileGroup11], [fileGroup12])

如果要刪除分割槽方案

--刪除分割槽方案
drop PARTITION SCHEME [PS_card]

(5).上面建立了2017年12個月的分割槽，如果要新增2018年12個月的分割槽並且複用fileGroup01~fileGroup12

--修改邊界值
ALTER PARTITION SCHEME [PS_card] NEXT USED [fileGroup01]
alter partition function [PF_card]()  SPLIT RANGE ('2018-01-01')

ALTER PARTITION SCHEME [PS_card] NEXT USED [fileGroup02]
alter partition function [PF_card]()  SPLIT RANGE ('2018-02-01')

ALTER PARTITION SCHEME [PS_card] NEXT USED [fileGroup03]
alter partition function [PF_card]()  SPLIT RANGE ('2018-03-01')

ALTER PARTITION SCHEME [PS_card] NEXT USED [fileGroup04]
alter partition function [PF_card]()  SPLIT RANGE ('2018-04-01')

ALTER PARTITION SCHEME [PS_card] NEXT USED [fileGroup05]
alter partition function [PF_card]()  SPLIT RANGE ('2018-05-01')

ALTER PARTITION SCHEME [PS_card] NEXT USED [fileGroup06]
alter partition function [PF_card]()  SPLIT RANGE ('2018-06-01')

ALTER PARTITION SCHEME [PS_card] NEXT USED [fileGroup07]
alter partition function [PF_card]()  SPLIT RANGE ('2018-07-01')

ALTER PARTITION SCHEME [PS_card] NEXT USED [fileGroup08]
alter partition function [PF_card]()  SPLIT RANGE ('2018-08-01')

ALTER PARTITION SCHEME [PS_card] NEXT USED [fileGroup09]
alter partition function [PF_card]()  SPLIT RANGE ('2018-09-01')

ALTER PARTITION SCHEME [PS_card] NEXT USED [fileGroup10]
alter partition function [PF_card]()  SPLIT RANGE ('2018-10-01')

ALTER PARTITION SCHEME [PS_card] NEXT USED [fileGroup11]
alter partition function [PF_card]()  SPLIT RANGE ('2018-11-01')

ALTER PARTITION SCHEME [PS_card] NEXT USED [fileGroup12]
alter partition function [PF_card]()  SPLIT RANGE ('2018-12-01')

(6) .按分割槽建立表和索引

create table gf_card (
  id            bigint not null,
  crt_time        datetime not null,
  name            varchar(20) not null
) on [PS_card] (crt_time)

對於分割槽表，如果要建主鍵，必須把分割槽依據列crt_time包含在內，同時如果要建唯一索引也要把分割槽依據列crt_time包含在內，所以分割槽表不能再單獨對id建主鍵，也不能單獨對id建唯一索引，但可以建非唯一索引

-- 將索引放在[PRIMARY]檔案組中
CREATE CLUSTERED INDEX [IX_card_id] ON gf_card (id) ON  [PRIMARY]

由於crt_time查詢時經常會用到，所以在crt_time上也建索引，並且也將各分割槽的索引放入各分割槽中

CREATE NONCLUSTERED INDEX [IX_card_crt_time] ON gf_card (crt_time) on [PS_card] (crt_time)

(7).插入一些資料測試

insert into gf_card (id, crt_time, name) values (99, '2017-07-01', 'n99')
insert into gf_card (id, crt_time, name) values (100, '2017-07-01', 'n100')
insert into gf_card (id, crt_time, name) values (101, '2017-07-01', 'n101')
insert into gf_card (id, crt_time, name) values (200, '2017-07-02', 'n102')
insert into gf_card (id, crt_time, name) values (600, '2017-07-01', 'n600')

查詢表中資料在哪個分割槽上

select *, $partition.PF_card(crt_time) partitionNum from gf_card

id        crt_time                                   name    partitionNum
99      2017-01-01 00:00:00.000    n99        2
100    2017-07-01 00:00:00.000    n100     8
101    2017-11-01 00:00:00.000    n101     12
200    2018-01-02 00:00:00.000    n102     14
600    2018-07-01 00:00:00.000    n600      20
從結果中可以看到fileGroup01的分割槽號為2，fileGroup07的分割槽號為8，primary的分割槽號自然就為1 了

--查詢資料在哪個分割槽上
select $partition.[PF_card]('2017-07-01')

--查詢表中資料在哪個分割槽上
select $partition.[PF_card](id)
from gf_card

-- 查詢表有哪些分割槽
SELECT *
FROM sys.partitions AS p JOIN sys.tables AS t  ON  p.object_id = t.object_id
  WHERE p.partition_id IS NOT NULL
      AND t.name = 'gf_card';

(8).將分割槽中的資料刪除

比如只刪除fileGroup01中的資料，由於直接使用truncate刪除分割槽中的資料要到SQL server2016才支援，所以使用一種通用的方法來刪除。
建立一個與gf_card列相同的表gf_card_tmp，並且此表也要建在fileGroup01中

create table gf_card_tmp (
  id              bigint not null,
  crt_time        datetime not null,
  name            varchar(20) not null
) on [fileGroup01]

將分割槽2中的資料轉移到gf_card_tmp中，此時gf_card分割槽2中就沒有資料了

alter table gf_card switch partition 2 to gf_card_tmp

刪除gf_card_tmp中的資料

truncate table gf_card_tmp

(9).如何查詢庫中有哪些PartitionScheme和PartitionFunction？

select ps.Name PartitionScheme, pf.name PartitionFunction
 from sys.indexes i
 join sys.partition_schemes ps on ps.data_space_id = i.data_space_id
 join sys.partition_functions pf on pf.function_id = ps.function_id
where i.object_id = object_id('gf_card')