Is it OK to extend the CPU core to Elasticsearch?


(Shota Ito) #1

I am planning to introduce the Elasticsearch server in earnest.
We plan to use an on-premises server due to security restrictions.
So I thought.
How far is the Elasticsearch's CPU processing to load balance nicely?
Normally, excessive multicore thinks whether it causes context switch.

I also understand the upper limit of JVM's object pointer.

How much core do you have on one server?
Note: Because I plan to use high-performance SSD, Network thinks I will become a bottleneck before DISK I/O.

By the way, as far as the budget permits, considering the hardware performance it has the following structure.

CPU: Intel Xeon E5-2698 * 2 (Total 80 threads)
SSD: Intel SSD 750
Memory: 128 GB

If the CPU becomes a bottleneck on the contrary,
I'm thinking to divide it by KVM, Docker, etc and use it.
I am happy to be able to use Elastic Cloud Enterprise as soon as possible :slight_smile:


(system) #2

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.