Moving TempDB

本文介绍如何通过内部修改sys.database_files来移动SQL Server中的TempDB位置,包括具体的T-SQL语句及重启服务后的文件变化情况。

By Andy Warren, 2008/08/04

Moving TempDB

Recently I had a question about moving TempDB, it didn't behave as the user had anticipated. TempDB is a special case, so let's look at how to move it and what happens when the service is restarted. Our first thought on moving a database is to detach, cut/paste, reattach. Works well enough for normal databases but not the system ones (system=we need them all the time). Plan B is to do a backup, then a restore with move. Again, that works for normal databases, but that doesn't solve our problem.

So finally we look it up, here's the code needed to move TempDB to C:/:

USE master;
GO
ALTER DATABASE tempdb
MODIFY FILE (NAME = tempdev, FILENAME = 'c:/tempdb.mdf');
GO
ALTER DATABASE tempdb
MODIFY FILE (NAME = templog, FILENAME = 'c:/templog.ldf')
 

If you think about what that does internally, we're really modifying sys.database_files. Run the query select * from tempdb.sys.database_files to see the files for any database. For my current tempdb, it returns:

 

Zoom out   |   Open in new window

 

If I run the ALTER from above, I get this message:

 

Zoom out   |   Open in new window

 

And if I re-run the select, I get this:

 

Zoom out   |   Open in new window

 

What's unusual about that is that it shows the files being in the new location, but if you look on disk there are no files in the new location, they are still at the old location! Everything works. So how do the files get to the new location, and when? If you execute the script and then stop the service, you'll see that TempDB still exists in it's old location and does not exist at the new location. When the service is restarted TempDB will be created in it's new location, but the old TempDB files will still exist at the original location. The old TempDB files can be deleted at any time once the service has been stopped.

I went a step further and stopped the service again, renamed the two TempDB files (just in case!) and restarted the service without error. So we can see that we get a new TempDB if we execute a move, or delete a file, but what happens on just a plain service restart? I executed the following query in TempDB:

create table AndyTest (SomeKey int)

After restarting the service, did the table still exist? No, I get an invalid object name error. The only way to add objects to TempDB that will persist across a service restart is to add the objects to Model, or to execute a startup script that will recreate them (I like the latter approach better if you have to do this).

Moving TempDB isn't something we'll do often, but at least it's about as painless as you can ask for once you know the secret syntax.

Visit my blog at http://blogs.sqlservercentral.com/andy_warren/default.aspx

By Andy Warren, 2008/08/04

内容概要:本文介绍了一种用于电磁暂态(EMT)研究的第四类全变流器型风力发电系统的通用Simulink仿真模型,旨在构建一个能够准确反映实际风电系统动态特性的简化通用模型。该模型涵盖了风力机、传动链、发电机、全功率变流器及其控制策略等关键组成部分,重点突出系统在电网故障、风速波动等复杂工况下的动态响应能力,适用于风电并网电磁暂态分析、新型电力系统稳定性评估及高比例可再生能源接入场景的研究。模型设计兼顾准确性与仿真效率,便于研究人员快速搭建和调试,推动风电系统建模与控制技术的发展; 适合人群:具备一定电力系统理论基础和MATLAB/Simulink仿真能力,从事新能源发电、电力电子变换、风电并网控制及相关方向的研究生、科研人员及工程技术人员; 使用场景及目标:①开展风电系统在电网扰动下的电磁暂态仿真分析;②研究全功率变流器风电机组的动态行为与控制特性;③支撑新型电力系统中高渗透率风电接入的稳定性与电能质量评估,服务于学术研究、课程教学与工程项目前期仿真验证; 阅读建议:建议读者结合文中提供的模型结构与参数设置,在Simulink环境中动手复现并调试仿真模型,通过设置不同运行工况(如三相短路、低电压穿越、风速突变等)观察系统响应,深入理解全变流器风电机组的建模方法、控制逻辑与动态特性,进而拓展应用于更复杂的多机并网或综合能源系统仿真场景。
内容概要:本文探讨了Hash算法在芯片行业,特别是网络处理器(NPU)中的性能优化实践,聚焦于如何通过硬件流水线设计实现纳秒级高速哈希查找。文章分析了传统软件哈希方案在高带宽场景下的局限性,提出基于硬件描述语言或高层次综合(HLS)的设计方法,采用CRC32等硬件友好型哈希函数、并行多路哈希、流水线化解耦以及双端口存储结构等核心技术,实现低延迟、无阻塞的数据包处理。通过C++风格的HLS代码示例,展示了从算法选择到资源优化的完整硬件加速逻辑,并深入剖析了循环展开、位宽控制、内存映射和流水线调度等关键实现细节。最后展望了可编程交换机、P4语言支持及TCAM与哈希混合架构等未来发展方向。; 适合人群:具备数字电路基础、熟悉Verilog/SystemVerilog或HLS的芯片设计工程师,以及从事网络芯片、FPGA加速、高性能路由设备研发的技术人员(工作年限1-5年为宜);也适合对硬件加速算法感兴趣的研究人员。; 使用场景及目标:①掌握如何将哈希算法高效映射到硬件逻辑中以满足线速转发需求;②理解流水线设计、资源约束优化、读写冲突规避等在实际芯片项目中的应用;③为开发高性能网络处理器中的流表/路由表查找单元提供技术参考与实现范例。; 阅读建议:学习时应结合HLS工具(如Xilinx Vitis HLS)进行代码仿真与综合,重点关注#pragma指令对硬件结构的影响,并对比不同哈希算法在资源占用与时序表现上的差异,深入理解“空间换时间”的硬件优化本质。
内容概要:本文系统研究了基于粒子群优化(PSO)、灰狼优化(GWO)、鲸鱼优化(WOA)、哈里斯鹰优化(HHO)、蜣螂优化(DBO)和麻雀搜索算法(SSA)六种智能优化算法在无人机三维路径规划中的应用,并在Matlab平台上实现了相应的仿真代码。研究构建了复杂的三维地形与障碍物环境模型,设计了包含路径长度、飞行时间、能耗及安全性等多维度的成本函数,通过定义合理的适应度函数与飞行约束条件,对各类算法的路径搜索能力、收敛速度与规划质量进行了全面对比分析。通过仿真实验验证了各算法在静态复杂环境下的性能表现,旨在为实际无人机任务中优化算法的选择提供科学依据和技术支持。; 适合人群:具备一定Matlab编程基础,从事智能优化算法、无人机路径规划、自动化控制及相关领域研究的研究生、科研人员或工程技术人员。; 使用场景及目标:① 掌握主流群智能优化算法在无人机三维路径规划中的建模与实现方法;② 对比分析不同算法在相同复杂环境下的性能差异,为算法选型提供量化依据;③ 为后续研究动态环境路径规划、多无人机协同任务分配等问题奠定理论与技术基础; 阅读建议:建议读者结合提供的Matlab代码逐模块分析算法实现流程,重点关注适应度函数的设计、障碍物规避策略的实现以及多目标成本函数的权衡机制,同时可通过调整参数设置或引入新型优化算法进行扩展实验,以深化对算法性能的理解与实际应用能力。
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值