Server Fault Asked by The-Big-K on December 3, 2021
Does dedicated instances have better latency between nodes in the same AZ? Can someone comment on the inter-node latency for the following case:
1) SharedVPC.AZ1.instance1 → SharedVPC.AZ1.instance2 (same AZ, shared)
2) DedicatedVPC.AZ1.instance1 → DedicatedVPC.AZ1.instance2 (same AZ, dedicated)
The internals of Dedicated instances in EC2 is not shared in the documentation. It merely says that the hardware is not shared with other customers, but does not speak about connectivity differences between nodes.
I didn’t bother doing an actual experiment (yet) because, measuring a micro/milli second latency of a cloud provider’s instances differentiated by a small feature (dedicated tenancy) might lead to false conclusions, unless you do some low level OS tweaks (network card pinning, MTU size etc). I’m wondering if someone has prior experience with this data instead. Let me know! Thanks.
Dedicated tenancy is more intended for compliance reasons where you need to prove that no one else's workloads run on the same physical machine. It may or may not have speed benefits as a side effect but it's certainly not the primary reason for using Dedicated instances.
If you want to achieve low latency have a look at Placement Groups and in particular Cluster Placement Group that is designed for low latency:
Cluster – packs instances close together inside an Availability Zone. This strategy enables workloads to achieve the low-latency network performance necessary for tightly-coupled node-to-node communication that is typical of HPC applications.
Hope that helps :)
Answered by MLu on December 3, 2021
Get help from others!
Recent Questions
Recent Answers
© 2024 TransWikia.com. All rights reserved. Sites we Love: PCI Database, UKBizDB, Menu Kuliner, Sharing RPP