I have a MS SQL Server 2014 database with large scale of data an lots of users; and the RAID level is equal to 10. Now due to the rapid growth of data amount, I'm going to distribute this one database over two servers. What's the best solution for that?

migrated from stackoverflow.com Dec 1 '15 at 2:03

This question came from our site for professional and enthusiast programmers.

  • Why not just put all the ram and CPU on one server? What do you think you'll gain from two servers? Do you have a perforamance issue? Have you worked out what the bottleneck is? – Nick.McDermaid Nov 30 '15 at 13:12
  • also look into filegroups and/or table partitioning. if your data is growing large, you can put certain tables , indexes, or even parts of tables on their own disks. link to get you started: technet.microsoft.com/en-us/library/ms179316(v=sql.105).aspx – Jeremy Nov 30 '15 at 15:47
  • 1
    It depends on too many variables. You mentioned only the fact that you have a lot of data but we don't know anything about uptime requirements yet. In purely a data distribution view, you could get more space by adding another node by federating, or sharding your application. This is usually something expensive and requires a lot of forethought, even in a nosql solution like mongo where it's built in. You might need to scale up instead of out for this solution based on your needs and budget. Please clarify the answer. – Ali Razeghi Dec 1 '15 at 2:24

Try using sql server failover cluster

  • Failover clustering uses shared storage and gives no data volume benefits. It protects you against hardware failure of a node at the expense of a shared SAN. You could use 2012 AGs instead of failover clustering but it would just give you non shared storage. Clustering will not help scale with volume of data in any way, just with availability of services. – Ali Razeghi Dec 1 '15 at 2:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.