0

I have a VM instance which has the following configuration and runs the geth blockchain on it.

20 GB HDD (boot disk)
500 GB SSD (chain data)

The geth node works ok and is in sync. as expected. I want to put such instances behind a HTTP load balancer, however I am not sure of the best way to scale instances as it also includes chain data on the the mounted disk.

I think I have the following options when wanting to scale instances:

  1. Create an instance with Boot disk SSD (500GB) and create an image which can then be used to create an Instance Template. (The blockchain will still take time to sync.)

  2. Keep the chaindata in the mounted disk and create an image. When scaling instances, create a disk from this image and attach it to the newly launched instance using gcloud. (The blockchain will still take time to sync.)

Is there a better way to do it than any of the above approaches?

Maxim
  • 4,075
  • 1
  • 14
  • 23
kosta
  • 4,302
  • 10
  • 50
  • 104

1 Answers1

0

The better way to do it is with Google Kubernetes Engine VM nodes. It is better to run geth on GKE. This will speed up the sync process, as you can create a StatefulSet with auto-scaling and each pod will be able to have its own SSD disk.

For more information in this regard, you may follow the instructions in this thread.

Mahmoud Sharif
  • 1,085
  • 6
  • 12