Comments on: The Accidental DBA (Day 2 of 30): Hardware Selection: Disk Configurations and RAID -> Performance not Capacity

By: Victor B

Victor B — Sat, 22 Aug 2015 16:34:24 +0000

Hi Glenn.

First of all, thanks for the post and thanks to SQLSkills team for the entire 30-day series of post. When I read about performance in blogs or books I always read that we need to think in consideration of our workload type, but I have not read much about “workload types”, I mean, do you know any blog/book/whitepaper where I can read and learn about the recommendations on the different “workload types”? For example, I guess a reporting database should be configured in a different way than a point of sale system and different than a datawarehouse, etc. So, any recommendations on that matter will be highly appreciated.

Thanks! 🙂

By: Mark V

Mark V — Tue, 13 May 2014 02:41:11 +0000

Glenn,

How do you deal with the SAN volumes that are presented to DBA and you don’t have any visibility in SAN settings?

Do you know of any resources that explain SAN caching, volumes, storage pools, stripe sizes, etc?

I’d like to understand SAN settings better and also be able to educate the SAN admin 🙂

Thanks,
Mark

By: Kukah

Kukah — Thu, 13 Jun 2013 19:27:57 +0000

Understood.

Thank you.

By: Glenn Berry

Glenn Berry — Thu, 13 Jun 2013 13:06:42 +0000

In reply to Kukah. When a server is part of a fail-over cluster instance, the SQL Server binaries are usually installed on one of the cluster drives instead of the local C: drive.

By: Kukah

Kukah — Thu, 13 Jun 2013 13:00:01 +0000

Hi Glenn,

Thanks for the explanations.

Could you explain what do you mean with (for a server that is not part of a fail-over cluster instance), has we are just installing a SQL Cluster with RAID 1 for binaries ?

Is it for the ‘decent performance’ or for the ‘redundancy’ part ?

Thank you,
Kukah