November 21, 2017 at 9:44 am
We have a data warehouse (400GB) which is running on SQL Server 2008 standard edition. It is running on Windows server 2008R2. It has 2 processors and 32 GB of memory installed. The plan is to move that database to a newer hardware with SQL Server 2016 (Standard edition). We are still not sure whether to stay on Prem or move to AWS (whichever is cheaper). My thoughts were to have 2 servers, move the database, have AG on etc. My manager thinks it is best that we move it to AWS since it is easier, cheaper, and the data is much more secure. Is there like a documentation (prerequisites) on what to do or things to consider or how to best build a new server from scratch? What sort of guidelines do I have to follow in order for us to migrate the database? Should we stay on Prem or move to AWS? Pros and cons? Can I please get some advises from experts?
November 21, 2017 at 3:49 pm
Hi,
The Standard tier in Azure supports databases up to 1Tb in size. If your database will grow you may need to switch to the Premium tire which is much more expensive.
Standard 100DTU $168.83 per month
Premium 250DTU $915.00 per month
https://docs.microsoft.com/en-us/azure/sql-database/sql-database-service-tiers
You can use a calculator - https://azure.microsoft.com/en-us/pricing/calculator
November 21, 2017 at 4:56 pm
The non-premium rates have a charge for every 10,000 "transactions". They don't define what a transaction is, though, or at least I've not found it yet. If a "transaction" is a logical read, it would cost us a fortune. We have some procs that clear over a billion reads on a single run and they run a whole lot. Of course, we're in the process of fixing that mess but my advise would be to figure out what you're server is doing in the area of "transactions" according to whatever the "transaction" is for AWS, which could become an acronym for what your boss will say when he get's the bill (AW S***!!!) 😉
--Jeff Moden
Change is inevitable... Change for the better is not.
November 22, 2017 at 10:07 am
I worked with a company which had their entire infrastructure on AWS when the DDOS attack took out AWS, they (Amazon) may have solved the issue of potential attacks, I am not sure; I had to prove that the responsibility for loss of service laid at their door and not the IT team door. Their S2 solution has been suspect even recently. If you go for AWS do not use their fast drives (they are ephemeral) for any critical database as the drive disappears if server restarts! Dead handy that! I may be prejudiced but these are issues you need to be aware of IMHO. The company has since reverted to tin.
...
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply