Examine Best-Practices to Run LLMs on Host

[doronkg]

Examine the following methods of deploying the LLM model on a host:

1. Container
2. Guest OS

Consider the following aspects:

• Resource utilization overhead/latency (GPU, CPU, Memory, Disk, etc)
• Host compatibility (OS, processor type, etc)