What are the important factors when selecting hardware and OS for a super fast EOS Node.
I'm assuming single threaded processors only count? A boat load of RAM? Some OS tweaking?
I really want to understand the factors that drive performance.
Add a comment
|
1 Answer
You can check the performance of EOS New York (they are no more one of the 21 BP, but they have been for a long time)
Core Layer
One producing node and multiple fully redundant backups in a fail-over model comprise our core layer. The main producing node is the workhorse of the operation. A backup is activated during system upgrades.
80 GBs - 3 TBs of RAM (as needed)
4-8 vCPUs
1 TB of SSD Storage
As you can see the most important resources is RAM with a gigantic amount of 3TB, for the OS as long as you can run docker I think you are fine but most BP tend to pick linux
check this question for more information, there is lots of link for articles about the hardware and configuration of BP What is the system requirements of being a block producer?
-
Do the vCPU's add anything as eosio is single threaded? Is there any particular flavor of linux or any modifications to the basic OS install that can help the need for speed?– BillCommented Sep 13, 2019 at 16:25
-
Currently it is single thread, but in the future especially with the 1.8 version coming this month they plan to try to multithread some of the stuff, for the OS part maybe you can get marginal gain by tweaking the server side but a recent ubuntu will be fine I guess Commented Sep 13, 2019 at 17:46