我在 sql 服务器中的搜索速度很慢

Question

我有一个 table，你可以在我的 sql 服务器中看到它：

CREATE TABLE [dbo].[Cars](
    [Id] [bigint] IDENTITY(1,1) NOT NULL,
    [VIN] [nvarchar](max) NULL,
    [ChassisNumber] [nvarchar](max) NULL,
    [PlaqueCoded] [nvarchar](max) NULL,
    [EngineNumber] [nvarchar](max) NULL,
    [SystemType] [nvarchar](max) NULL,
    [CarType] [nvarchar](max) NULL,
    [CarTipe] [nvarchar](max) NULL,
    [FuelType] [nvarchar](max) NULL,
    [FuelSystem] [nvarchar](max) NULL,
    [Model] [int] NULL,
    [Color] [nvarchar](max) NULL,
    [SubmitDatetime] [datetime] NOT NULL,
    [ExpireDatetime] [datetime] NOT NULL,
    [ReferenceOrganization] [nvarchar](max) NULL,
    [ReferenceId] [nvarchar](max) NULL,
    [Comment] [nvarchar](max) NULL,
 CONSTRAINT [PK_Cars] PRIMARY KEY NONCLUSTERED 
(
    [Id] ASC
)WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON [PRIMARY]
) ON [FullOrderDateRangePScheme]([Model])

我在这个 table 里面有 16000000 条记录。所以你知道这个 table 存储了关于 cars.so 的所有信息我决定创建基于车型的文件组，这意味着我保存了如您所见，同一个文件组中有相同型号的汽车：

 FILEGROUP [Filegroup_1395] 
( NAME = N'data_1395', FILENAME = N'C:\Program Files\Microsoft SQL Server\MSSQL13.MSSQLSERVER\MSSQL\DATA\data_1395.ndf' , SIZE = 10240KB , MAXSIZE = 10240000KB , FILEGROWTH = 512000KB ), 
 FILEGROUP [Filegroup_1396] 
( NAME = N'data_1396', FILENAME = N'C:\Program Files\Microsoft SQL Server\MSSQL13.MSSQLSERVER\MSSQL\DATA\data_1396.ndf' , SIZE = 10240KB , MAXSIZE = 10240000KB , FILEGROWTH = 512000KB ), 
 FILEGROUP [Filegroup_1397] 
( NAME = N'data_1397', FILENAME = N'C:\Program Files\Microsoft SQL Server\MSSQL13.MSSQLSERVER\MSSQL\DATA\data_1397.ndf' , SIZE = 566976KB , MAXSIZE = 10240000KB , FILEGROWTH = 512000KB )

所以我创建这个函数是为了

CREATE PARTITION FUNCTION [FullOrderDateKeyRangePFN](int) AS RANGE LEFT FOR VALUES ( 1395, 1396, 1397)
GO
/****** Object:  PartitionScheme [FullOrderDateRangePScheme]    Script Date: 8/8/2017 11:51:38 PM ******/
CREATE PARTITION SCHEME [FullOrderDateRangePScheme] AS PARTITION [FullOrderDateKeyRangePFN] TO ( [Filegroup_1395], [Filegroup_1396], [Filegroup_1397])
GO

请注意，VIN 是唯一的。我的主要搜索是 vin.i 在我的 table 上创建这些索引：

CREATE CLUSTERED INDEX [ClusteredIndex-20170808-232559] ON [dbo].[Cars]
(
    [Model] ASC
)WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, SORT_IN_TEMPDB = OFF, DROP_EXISTING = OFF, ONLINE = OFF, ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON [FullOrderDateRangePScheme]([Model])
GO
SET ANSI_PADDING ON
GO

CREATE NONCLUSTERED INDEX [NonClusteredIndex-20170808-232524] ON [dbo].[Cars]
(
    [Id] ASC
)
INCLUDE (   [VIN]) WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, SORT_IN_TEMPDB = OFF, DROP_EXISTING = OFF, ONLINE = OFF, ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON [FullOrderDateRangePScheme]([Model])
GO

我的查询是这样的：

select * from cars where vin='IRFC1374GH7162K'

我的问题是：
我怎样才能做更多的事情来提高性能？我的索引会有帮助吗？

我是调音新手

Answer 1

你的查询是这个

select * from cars where vin='IRFC1374GH7162K'

您当前的索引对您的查询完全没有帮助..

为了让您的查询执行得更好，我建议..尝试添加一个非聚集索引

CREATE NONCLUSTERED INDEX [NonClusteredIndex-20170808-232524] ON [dbo].[Cars]
(
    [vin] ASC
)
INCLUDE (  remaining columns)

这将是一个问题，因为如果您使用的是 sql2016 之前的版本，则最大索引键大小为 900 字节，对于大于等于 2016 的版本，限制为 1700 字节..

我还看到，您的搜索列不是那么大（IRFC1374GH7162K），所以我建议调整为 nvarchar(100)。此外，所有列都是必需的，因为您正在执行 select *

有了这个，您将有维护一个单独结构的开销...但这有助于您的查询..如果您不需要所有列，那么开销会更少

我在 sql 服务器中的搜索速度很慢

My search is very slow in sql server

sql-server

indexing

performance

database-tuning