Chrona Workers

Cloud-hosted worker infrastructure for running your agents

What are Chrona Workers?

Chrona is the execution plane for agent tasks. Cloud worker pools (below) are hosted and scaled by Viksa; when you set an agent to Secure, the same engine dispatches to workers you run in your environment instead.

Each organization gets a dedicated worker cluster that can be scaled horizontally (multiple replicas) and vertically (V-Series resource profiles) based on your workload requirements and plan limits.

Cloud vs Secure Workers

Cloud Workers

  • • Fully managed by ViksaAI
  • • Instant horizontal scaling
  • • V-Series resource profiles
  • • Zero infrastructure setup
  • • Subject to plan limits

Secure Workers

  • • Run on your infrastructure
  • • Full control over resources
  • • Access to private networks
  • • Custom environment setup
  • • No plan limits apply

Key Features

V-Series Profiles

Choose from resource tiers (Glide through Nova) for optimal performance.

Horizontal Scaling

Scale replicas up or down based on workload demand.

Pause & Resume

Pause workers to save resources, resume when needed.

Live Logs

View real-time logs for each pod in your cluster.

Usage Tracking

Monitor CPU and memory usage against plan limits.

Auto Updates

Worker images are automatically updated with latest features.

V-Series Resource Profiles

Each V-Series profile provides a different combination of CPU and memory resources. Choose the profile that matches your workload requirements.

ProfileCPUMemoryBest For
V-Glide100m128MiGlide — entry tier
V-Dash200m256MiDash — everyday load
V-Pulse250m256MiPulse — compact rhythm
V-Bolt300m384MiBolt — extra codegen room
V-Cadence500m512MiCadence — default balance
V-Storm750m768MiStorm — higher concurrency
V-Nova1000m1GiNova — flagship throughput

Usage & Plan Limits

Cloud worker resources are subject to your plan limits. Usage is calculated as:

Total Usage = Replicas × Per-Replica Resources

For example, if you have 2 replicas with V-Cadence profile (500m CPU, 512Mi memory), your total usage is 1000m CPU and 1024Mi memory. The UI displays your current usage against your plan limit (e.g., "1000 / 2000m").

Worker Lifecycle

Running

Workers are active and processing tasks

Scaling

Pods are being added or removed

Paused

Scaled to 0 replicas, no resources consumed

Related

Platform overview, Cloud vs Secure agents, Core concepts