从SQL Server 2005中的单个表进行数据访问太慢

Question

Following is the script of table. Accessing data from this table is too slow.

SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
CREATE TABLE [dbo].[Emails](
    [id] [int] IDENTITY(1,1) NOT NULL,
    [datecreated] [datetime] NULL CONSTRAINT [DF_Emails_datecreated]  
        DEFAULT (getdate()),
    [UID] [nvarchar](250) COLLATE Latin1_General_CI_AS NULL,
    [From] [nvarchar](100) COLLATE Latin1_General_CI_AS NULL,
    [To] [nvarchar](100) COLLATE Latin1_General_CI_AS NULL,
    [Subject] [nvarchar](max) COLLATE Latin1_General_CI_AS NULL,
    [Body] [nvarchar](max) COLLATE Latin1_General_CI_AS NULL,
    [HTML] [nvarchar](max) COLLATE Latin1_General_CI_AS NULL,
    [AttachmentCount] [int] NULL,
    [Dated] [datetime] NULL
) ON [PRIMARY]

Following query takes 50 seconds to fetch data.

select id, datecreated, UID, [From], [To], Subject, AttachmentCount, 
    Dated from emails

If I include Body and Html in select then time is event worse.

indexes are on:

• id unique clustered
• From Non unique non clustered
• To Non unique non clustered

Tabls has currently 180000+ records.

There might be 100,000 records each month so this will become more slow as time will pass.

Does splitting data into two table will solve the problem? What other indexes should be there?

Answer 1

It's almost certainly the volume of the data that's causing a problem. Because of this, you should not fetch the Subject column until you are need it. Even fetching SUBSTRING(Subject, 100) may be noticeably faster.

This may be irrelevant, but older versions of SQL Server suffered if the BLOB columns weren't the last in the row, so just as an experiment I'd move [AttachmentCount] and [Dated] above the three nvarchar(max) columns.