1 Answer
- Newest
- Most votes
- Most comments
0
Hi - Thanks for reaching out. I think below resources should help a lot
- https://docs.aws.amazon.com/prescriptive-guidance/latest/sql-server-ec2-ha-dr/single-node.html#instance-type - this does talks about some of the proposed instance types
- https://docs.aws.amazon.com/prescriptive-guidance/latest/sql-server-ec2-best-practices/welcome.html - this should be looked into example Avoid CPU core mismatches etc.
Thanks
Relevant content
- asked 2 years ago
- asked 2 months ago
- asked 2 months ago
- AWS OFFICIALUpdated 2 years ago
- AWS OFFICIALUpdated 3 years ago
- How can I capture information about a deadlock on my Amazon RDS DB instance that runs on SQL Server?AWS OFFICIALUpdated 5 months ago
- AWS OFFICIALUpdated 5 months ago
Thanks for the response.
I'm bit not clear on how many vCPUs should be required for this. The Intel Xeon Gold on-remises processor has 20 cores. And there are two such processors on that host. So total of 40 cores.
So should I select an instance with at least 80 vCPUs and 40 cores? Like an r5.24xlarge instance with 48 cores and 96 vCPUs?
One of the benefits of the cloud is agility - when you purchased those Xeon Gold processors, you most likely had to use them until your organization went through another refresh cycle. With AWS, you can pick an instance (such as the R5.24xlarge), get a performance profile of your workload, and, if the instance is oversized or undersized, easily change what you're using. I'd look at a newer generation (i.e. R6i) if your workload is memory intensive. If your workload is compute intensive, you can check out the C series. Here's the link to how to change the instance type: https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ec2-instance-resize.html