December 6, 2013 at 8:21 am
Hello All,
we are planning to migrate SQL Server 2005 (Active-Passive Cluster) DW environment to SQL Server 2012 (Active-Passive with AlwaysOn). Below is the
current configuration of our SQL Server 2005 cluster instance.
SQL ClusterName : BWPROD
SQL Instance Name: BWPROD\BWSQL
OS : Windows server 2003 R2
IP Addresses
Cluster - xx.xx.xx.10
DTC - xx.xx.xx.11
SQL - xx.xx.xx.12
Server Configuration
ServerNameNICIP(Public)NICIP(Heartbeat)configurationmemoryno of processors
BWPROD1 xx.xx.xx.8xxx.xxx.x.1 128GB 24
BWPROD2 xx.xx.xx.9xxx.xxx.x.2 128GB 24
Drive Configuration
DriveLetter Label Total Size
C:\ localDisk 135 GB
D:\ Pagefile 80 GB shared drive
E:\ Data 11 TB shared drive
L:\ Logs 2 TB shared drive
M:\ MSDTC 500 MB shared drive
Q:\ Quorum 500 MB shared drive
T:\ TempDB 810 GB shared drive
As far as I understand we can create a WSFC without the need of any shared storage in order to use the alwayson feature in SQL Server2012. So this means I guess 2 stand alone servers with the same set of drive capacity on both the nodes BWPROD1 and BWPROD2. I understand this is expensive for our company and this takes a lot of explanation to do with the higher management and convincing them to go for this. I need few suggestions from SQL gurus here how to push this so that the management team can consider atleast.
1. Are there any alternatives in reducing the cost?
2. how to better configure a VM box for a secondary replica? ( There might be performance issues as we had it in the past)
3. if implemented, how do we install SQL patches or hotfixes? Any procedure available? Is it the same way as we do for a Active - passive cluster?
4. Please list all advantages of having AlwaysOn feature.
I really appreciate it if you can provide me any valuable suggestions in helping us implementing this in our production environment.
Thanks,
Srikanth
Thanks,
Srikanth
December 6, 2013 at 10:43 am
I would start by asking you what the main purpose of your setup is. Why do you want high availability? Are you protecting from site failure? Server Failure? Disk failure? Network Failure? High Availability Groups use what is essentially mirroring. No shared storage. Clustering will be the same as you are used to. Availability groups require Enterprise, whereas a failover cluster can use standard... Many more questions to answer. Start with what the ultimate goal is for your business. What are you selling to your stakeholders?
Jared
CE - Microsoft
December 9, 2013 at 7:13 am
Hello Jared,
How are you and thanks for the reply. I have created a document based on your questions of why we need to implement this and will submit it. At least this will help us in creating a test environment for further testing and answer the questions.
Thanks very much,
Srikanth
December 9, 2013 at 11:17 am
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply