SQL Server, SANs and Storage Tips

Featured Article(s)
I Caused a Security Breach…?
The President called me into his office not a month later and let me know that the CFO was no longer with the company. I was to shut down all his accounts and make sure that we recovered his laptop.

Troubleshooting Disk Performance
The disk system is a common source of performance issues on SQL Server. This often leads to finger pointing between disk administrators, DBA’s and developers as to where the bottlenecks are originating. SQL Sentry’s patented Disk Activity view gives you a complete real-time and historical display of your disk system as Windows sees it and pinpoints where those disk bottlenecks are occurring down to the data or log file level. Spot real time bottlenecks as well as historical trends and determine where the issue really originates. Get more information here.

SQL Server, SANs and Storage Tips
We’ve been talking a lot here about SANs and what you can expect, the pros, cons and so-on. We also have a webcast available on-demand in the video library. Check it out when you get a minute:

On-Demand Webcast: Understanding Key Fundamentals of SANs and SQL Server

Learn about some key returns on investment that you can expect with SANs, along with best practices, tips for best usage and how SQL Server and SANs can provide a compelling environment both in terms of availability and performance.

Trevor’s thoughts: "One of the things I have noticed is that the setup of the SAN can have a major impact on the performance of not only SQL but many other applications as well. Think and plan carefully the disk layout to ensure that your I/O patterns are appropriate to your systems. Putting heavy I/O systems on the same disk could cause unintended and hard to find issues with performance."

Jason: "I hear a common theme which at least 1 of the respondents took time to address. Just because you have Storage admins, does not mean you are not responsible for understanding how your application works. The best way to communicate to your Storage Admin would be to provide him an IO requirement instead of just a space requirement. As much as you may want to dream about different applications having exclusive storage, that will not always be the case. So, if an application admin could take a few moments out of his/her day to add a very small step to their process, you could avoid most problems.

That small step would be to capture disk IO usage and the scale of the environment. If the app admin then extrapolated that out by number of users or however the application load grows, then you could reasonably request a certain amount of IO to go along with the size. Assuming the storage admin is worth his paycheck, he will be able to associate that to an appropriate number of spindles (before SAN caching) and provide the application owner the resource. Ultimately, If an application owner understood the IO pattern of the application themselves, they would be better prepared to communicate to the storage people and (being the SQL admin as well here) it is appropriate that one of the parties be able to communicate to the lowest common denominator."

SQL Server Show
SelectViews: Today on the show we have three interviews – Kevin Kline, Scott Golightly and Anthony D’Angelo. We talked about all sorts of things, from Business Intelligence tricks to upcoming tools.
[Watch the Show]

Featured White Paper(s)
Scale Out SQL Server and Eliminate Outages with GRIDSCALE Database Virtualization Software
This technical white paper illustrates how you can leverage active-active database clustering technology to deliver scale-out… (read more)