请看下面的两个精彩解答:

解答1:

If your SAN has performance and availability algorithms built into the management software then yes, it is a good idea for both performance and availibility to put them on the SAN. As I understand it, SANs manage the pools and dynamically allocate storage as needed for either performance or disk failures (it senses that the disk is about to fail due to low performance metrics and physically moves files to other disks in the pool).

It is best practice to put log files and data files on seperate "physical" disks for 2 reasons:

  • First, your logfile is used as part of data recovery if your data files become corrupted.
  • Second, the logfile is read and write intensive because all DB activity (except for SELECT statements) are first written to the logfile, then out to data files. If you follow the sequence you would see that for a single transaction, it would have to (at a minimum) read from the same disk 4 times if logfiles and data files resided on the same disk. On seperate disks, it cuts the IO in half.

In an ideal situation, the log files and data files would reside on disks found on seperate controllers as this is the best performance and HA gain along with the disk speed and RAID choice.

 

解答2:

One thing to consider is that the log files are sequential writes where as the data files are non sequential. That is one of the reasons for separate LUNs. Log files write faster if they are on their own LUN because the spindles don't have to skip around, just write sequential. If you add in a data file then the spindles have to skip around and you lose some performance. I'm hoping I got the right terminology there as I'm not all that familiar with SANs themselves. The idea behind it should be however.

Frequently vendor recommendations are wrong when it comes to SQL Server. Just because SQL Server has different needs than most applications that use a SAN.

 

参考资料

=======================

SQL Server - Separating data, log, and TempDB files on a SAN

http://dba.stackexchange.com/questions/35404/sql-server-separating-data-log-and-tempdb-files-on-a-san

Any benefit to separate LUNs for log/DB when using SAN pooled storage?

http://community.spiceworks.com/topic/284204-any-benefit-to-separate-luns-for-log-db-when-using-san-w-pooled-storage?page=1#entry-1846277

是否应该将SAN上的SQL Server中的user database的data文件, log文件和TempDB文件放在不同的LUN上?的更多相关文章

  1. Linux上的SQL Server的起步

    我们知道,几个星期前,微软发布了在Linux上直接运行的SQL Server第一个公开CTP版本!因此,对我来说,是时候跨界在Linux上安装我的第一个SQL安装,这样的话,我就可以在Linux上折腾 ...

  2. Linux 上的 SQL Server 2017 的安装指南

    一:介绍背景 微软在2016年 3 月首次对外宣布了 Linux 版的 SQL Server,并于2017年 7 月发布了首个公开 RC 版.前几日在美国奥兰多召开的微软 Ignite 2017 大会 ...

  3. SQL Server 中 ROWLOCK 行级锁

    一.ROWLOCK的使用 1.ROWLOCK行级锁确保,在用户取得被更新的行,到该行进行更新,这段时间内不被其它用户所修改.因而行级锁即可保证数据的一致性,又能提高数据操作的并发性. 2.ROWLOC ...

  4. 谈一谈SQL Server中的执行计划缓存(上)

    简介 我们平时所写的SQL语句本质只是获取数据的逻辑,而不是获取数据的物理路径.当我们写的SQL语句传到SQL Server的时候,查询分析器会将语句依次进行解析(Parse).绑定(Bind).查询 ...

  5. 如何在64位系统上安装SQL Server 2000

    如何在64位系统上安装SQL Server 2000? 现在用SQL Server 2000数据库的人少了吧?大都是SQL Server 2005/2008了.不过还是有需求的,今天一朋友就让我在他的 ...

  6. SQL Server on Ubuntu——Ubuntu上的SQL Server(全截图)

    本文从零开始一步一步介绍如何在Ubuntu上搭建SQL Server 2017,包括安装系统.安装SQL等相关步骤和方法(仅供测试学习之用,基础篇). 一.   创建Ubuntu系统(Create U ...

  7. SQL Server on Red Hat Enterprise Linux——RHEL上的SQL Server(全截图)

    本文从零开始一步一步介绍如何在Red Hat Enterprise Linux上搭建SQL Server 2017,包括安装系统.安装SQL等相关步骤和方法(仅供测试学习之用,基础篇). 一.   创 ...

  8. Ubuntu上配置SQL Server Always On Availability Group(Configure Always On Availability Group for SQL Server on Ubuntu)

    下面简单介绍一下如何在Ubuntu上一步一步创建一个SQL Server AG(Always On Availability Group),以及配置过程中遇到的坑的填充方法. 目前在Linux上可以搭 ...

  9. Configure Always On Availability Group for SQL Server on RHEL——Red Hat Enterprise Linux上配置SQL Server Always On Availability Group

    下面简单介绍一下如何在Red Hat Enterprise Linux上一步一步创建一个SQL Server AG(Always On Availability Group),以及配置过程中遇到的坑的 ...

随机推荐

  1. 查看锁信息(开启InnoDB监控)

    当前mysql版本:5.6.21 一.背景 在mysql处理死锁问题时,由于show engine innodb status输出来的死锁日志无任务事务上下文,并不能很好地诊断相关事务所持有的所有锁信 ...

  2. HTML基础-DAY1

    HTML基础 Web的本质就是利用浏览器访问socket服务端,socket服务端收到请求回复数据提供给浏览器进行渲染显示. import socket def main(): sock = sock ...

  3. 转Google Protocol Buffer 的使用和原理

    Google Protocol Buffer 的使用和原理 Protocol Buffers 是一种轻便高效的结构化数据存储格式,可以用于结构化数据串行化,很适合做数据存储或 RPC 数据交换格式.它 ...

  4. android 单位 什么是屏幕密度?

    韩梦飞沙  韩亚飞  313134555@qq.com  yue31313  han_meng_fei_sha sp dp px in in 表示英寸, 是屏幕的物理尺寸.1英寸是2.54厘米. dp ...

  5. BZOJ 1061: [Noi2008]志愿者招募 [单纯形法]【学习笔记看另一篇吧】

    1061: [Noi2008]志愿者招募 Time Limit: 20 Sec  Memory Limit: 162 MBSubmit: 3975  Solved: 2421[Submit][Stat ...

  6. [LOJ#2980][THUSCH2017]大魔法师(线段树+矩阵)

    每个线段树维护一个行向量[A,B,C,len]分别是这个区间的A,B,C区间和与区间长度,转移显然. 以及此题卡常,稍微哪里写丑了就能100->45. #include<cstdio> ...

  7. bzoj 2850

    比较基础的KD树.每个节点维护一个BOX,包含包含当当前子树的点的最小矩形,以及点权和,然后用“整个矩形都在直线的一侧”和“整个矩形都不在直线的一侧”剪枝. /******************** ...

  8. Express浅析

    一.Express是什么? 首先Express是一个Node.js的Web框架,它的API使用各种HTTP实用程序方法和中间件,快速方便地创建强大的API,性能方面Express提供精简的基本web应 ...

  9. BZOJ 1192 鬼谷子的钱袋 数论

    1192:鬼谷子的钱袋 Time Limit: 10 Sec  Memory Limit: 162 MBSubmit: 1926  Solved: 1417 题目连接 http://www.lydsy ...

  10. wikioi 1294 全排列 dfs

    1294 全排列 时间限制: 1 s 空间限制: 128000 KB 题目等级 : 黄金 Gold       题目描述 Description 给出一个n, 请输出n的所有全排列 输入描述 Inpu ...