Hardware Requirements: Small-to-Mid Enterprise (Up to 100 Agents)
This profile is for production environments requiring stability and a minimum level of High Availability (HA) for business continuity.
⚙️ Deployment Prerequisites
See CX Deployment Prerequisites for detailed platform, operating system, and network prerequisites required before deploying Expertflow CX.
Core Platform (Mandatory)
The platform must be deployed with redundancy (HA) to prevent downtime of the unified routing and agent management functions. This typically requires a Kubernetes cluster with multiple nodes.
Component | Purpose | Min. Instances/Nodes | Recommended Specs (per VM/Node) | HA/DR Strategy |
CX-Core Cluster | Unified Routing, Agent Manager, Persistence, Analyzer, Basic Digital Channels | 3 Nodes (1 Control, 2 Worker) | 4 vCPU, 16 GB RAM, 250 GB SSD (≥10000 IOPS) | N+1 HA: Minimal redundancy for core services via distributed Kubernetes cluster. |
Data Pipeline Orchestrator | ETL and Data Integration (Apache Airflow) | 1 Worker Node | 4 vCPU, 8 GB RAM, 100 GB Disk | Mandatory Component: Requires a dedicated Worker Node for running scheduled data jobs. HA recommended. |
Databases | MongoDB, PostgreSQL (for CX-Core) | 3 Replicas (for data HA) | Resources shared/allocated from CX-Core Nodes | Databases run in a replicated set (e.g., MongoDB ReplicaSet) across the cluster nodes. |
Voice and Video Platform (If Voice is Required)
This sizing handles up to approximately 100 concurrent voice calls, with redundancy built-in.
Component | Purpose | Min. Instances/Services | Recommended Specs (per Instance) | Scaling/HA Considerations |
Media Server | Voice Processing, Recording, IVR | 2 System Services | 4 vCPU, 8 GB RAM, 500 GB SSD (≥10000 IOPS) | HA: Two identical, active servers for redundancy and load balancing. |
Jambonz (Voicebot) | Voice stream forking to ASR/TTS/NLU/LLM | 2 Sets of 3 VMs (HA) | As per PoC/Node role (e.g., 4 vCPU, 8 GB RAM) | Requires a redundant cluster setup (two instances of Feature/RTP/Node) for continuous bot operation. |
Add-on Components (Optional)
Component | Purpose | Min. Instances | Recommended Specs (per VM/Node) | Scaling Factor |
WFM (Workforce Management) | Workforce Management Backend/Frontend | 1-2 VMs/Nodes | 3 vCPU, 8 GB RAM, 50 GB SSD (≥10000 IOPS) | HA recommended. Sized for up to 100 agents. |
Surveys and Campaigns | Outbound Campaigning and Surveys Backend | 1 VM each | 2 vCPU, 4 GB RAM, 50 GB Disk | Can be co-located or deployed separately based on expected load. |
Third-Party AI/LLM | External Speech/Language Engines | Consult Vendor | Highly dependent on model. May require dedicated GPU. | Provision capacity based on peak expected transactions per second (TPS). |
💡 Summary for Presales
The standard requirement is a 3-node Kubernetes cluster for core services, complemented by two redundant Media Server instances for voice. Storage must be generously sized for voice recording retention.