Thank you Mr Ahmed for your time and you answer but still is more of what i was reading. Some forum said me that is not the load balance which we know, other said me is kind of difficult for a MSSQL Server, other say i need special hardware to accomplish.
I just want a concrect answer to this concret question, is possible make it work as real LB that 12 TB logistic "Disk" with a DB from MSSQL Server?
The meaning of my "Real Load Balance" is: one client send a workload process, that process convert in 4 segments to get to12 TB datastore cluster (which content a DB) through the 4 nodes (this way, get faster arrive, faster answer, and less bottle neck).
Again, thank you for time and sorry about my ignorance in this type of topic.
My best Regards!!!!!
Josh, please correct me if i'm wrong, You have 4 MS-SQL nodes and you need them to write/read at the same time to a shared 12 TB database.
If this is the case check the following documents :
The above covers the part of the diagram which includes the 4 SQL nodes and the shared database, The other part of the client to the SQL nodes which segments the loads on the nodes, i think you should consult Microsoft about the available options.
I've found some relevant links below:
Exactly Mr Ahmed!
Right Now i will check your links, then ill comment!
There is a name to what i want to do like "MSSQL DRS + HD installation"? or something like that? I ask this just for can google abt it better!.
Remember im really new on this topic and some terms or conceps dont handle it!
Mr ahmed, it looks like the hammer hit the nail :].
So i almost finish of been a trouble.
Last doubts, if the 4 mssql nodes are in a cluster with shared database and dont segments loads on nodes, where do it?
i thought was there, since the 4 nodes are on mode Load balance + HA + clustered and are working as 1 logical MSSQL SERVER which have 1 logical 12 TB disk (datastore cluster formed by each 3 TB virtual disk) or iam mixing everything?
In case iam wrong, I was thinking add an NBL (network balance load), i suppose in that case, the part of segments of my diagram will accomplish the goal.
Thank you again for your time and im sorry take your time!
My best Regards
Man, you are most welcome.
Regarding having the database distributed on 4 3TB LUNs each for a MS-SQL node, i think it could be supported by Microsoft to distribute the database on multiple disks.
Check this Microsoft document : Using Files and Filegroups
From my experience, i see many people deploy Microsoft clustering as follows :
A Microsoft cluster with 2 nodes, one active and one passive both are pointed to one or many shared physical RDM LUNs.
Regarding the NBL, I have found this document from Microsoft which states Network Load Balancing should not be used to directly scale applications, such as Microsoft SQL Server™ (other than for read-only database access), that independently update inter-client state because updates made on one cluster host will not be visible to other cluster hosts. To benefit from Network Load Balancing, applications must be designed to permit multiple instances to simultaneously access a shared database server that synchronizes updates. For example, Web servers with Active Server Pages should have their client updates pushed to a shared back-end database server.
Mr Ahmed, i think iam ready to my next step, thank to you.
Ill start with my lab to try all what we talk.
About 2 nodes, 1 active other passive, i read a lot abt it and how most of ppl use it but is jus HA and i want LB too.
Finally about NBL you gave an excellent tip, for sure i will fail on that if you dont tell that it cant work correctly with "write".
Really appreciated this chat!
Really thank you for share your knowlodge and experience!!
My best wish for you!
Have an excellent weekend!
You are welcome, please keep me posted with your progress.
Iam really new on this type of topic, so i apologize if iam doing or saying something incorrect. Im here to try to understand better this amaizing tool so i apreciate your knowlodge and time!!!!
Im trying to create a failover and load balancing cluster with Vsphere, the goal is try to install an MSSQL server DB where client can access more faster and not have downtime when a node is down.
I was reading a lot but still i cant have a full understand abt this, so if someone try to explain me something, please be patient and explain with details.
All this ill do it at home as a LAB, if really work and i can do it, ill try to use it for get a work since i dont have one and i really need one very soon!
Here is my diagram, is possible??
Well ill to explain it, i have clients which will send a workload process to my DB which will be allocate on MSSQL Server installed on logic disk of 12 TB created from the union of all virtuals storage. Each virtual storage is just a virtual disk created on a virtual machine which is hosted on physical server (node).
So i have 4 physical server which have 1 VM each one with 3 TB virtual disk, the goal is the client can send the workload process over the 4 nodes to have a faster response and if one node fail for some reason, the other node can work withow problem.
physical server S.O.: Dont know yet if i must use one or install VMWARE directly.
VM S.O: Windows Server 2008 (dont know if necessary or if i can just install SQL Server directly)
VM MSSQL Server: 2008 Enterprise (still dont know if i must use this one)
Well i can't think right now on other detail but if you point me ill be glad to write it.
sorry if my semantic or write is wrong but english is not my first lenguage.
Really thank you for get at this point, and share your knowlodge!!
My best Regards, josh from Argentine!!!