Linked Server SQL Server Optimization

Asked

Viewed 131 times

1

Currently has an environment that has 4 servers with database. I need to gather some information from 3 servers in 1.

Example:

The servers are: To B C and D

I need to get some information that’s on the servers To B C and place in the D.

My main concern now is with the optimization of this process, because it is a lot of data, the table of a database already has more than 1 million records.

Based on that password, using the Linked server would be best. The servers To B C send the information to D. Or the server D retrieve this data from other servers?

All databases are in SQL Server 2014.

DETAILS OF BELOW

  • The version of Sql Server is 13.0.1601.5;
  • Servers are all Windows Server 2012;
  • Are on the same network;
  • The server configuration is: Processor: Xeon E5 2GHz | 8 Gb ram memory | scsi disk;
  • And there are several other processes running in the database;

UPDATING

  • The SQL Server version is Entreprise;
  • I have access to sysadmin user;
  • 1

    Can you give more details about the Diego infra? How do A B C servers communicate with D ? Which version of SQL Server 2014 in each of these instances? Depending on your answer there are better solutions

  • This will be done only once?

  • This will run daily at night. Getting the new data from the tables. Only at the beginning it will be massive, because I have to get the data from 3 years ago.

  • In this scenario, Linked server can serve well. The first time will take time, but it will only be once. Then it will enter the routine and few data will be imported daily. Anyway, I would search D for A,B,C, a single point of maintenance.

  • @Cantoni really hadn’t thought about this maintenance issue yet.

  • Setting up re-plication in this case may be worth more. Version I mean what type of license (Standard, etc.) and whether you or dba are allowed to administer the instance.

  • @gmsantos. SQL Server is Enterprise. My user is sysadmin from all databases.

  • @Diegomoreno, if the problem is still pending, what is the criterion for taking information on servers A, B and C and transferring it to server D? There would be how on each source server (A, B and C), decide what to send to server D without having to query high volume of data from server D?

Show 3 more comments
No answers

Browser other questions tagged

You are not signed in. Login or sign up in order to post.