Optimal Management Method for a Two-Node Redundant Quantum Spark

[TSOL]

Hello Team

For managing a two-node redundant Spark setup, there are three available approaches:

1. HA Configuration
2. Spark Management via the Infinity Portal
3. Deploying Smart-1 On-premises on a VM

While using Spark Management via the Infinity Portal is generally recommended, the HA configuration appears to be the most cost-effective option.

My Question:
Which of these methods is considered the optimal management approach overall? Additionally, if we opt for the HA configuration due to its cost benefits, are there any notable disadvantages we should be aware of?

Thank you in advance.

[AkosBakos]

Hi @TSOL 

My opinon is to create a distributed setup with SMS (maybe on-prem). There are a lot of benefits, but I know this is more  expensive but it worth the money.

Akos

[G_W_Albrecht]

Your first point HA config means locally managed cluster, the other two centrally managed cluster. As every SMB has an included SMP (Spark Management Portal) license, i would opt for 2. as there is no price difference. This is preferable as:

- SMP Logging is superior (logs will vanish after a reboot in local management)

- firmware upgrades, backups and performing scripts on CLI is possible from the portal