Sizing Guide

Comprehensive sizing guidance for NeuronDB deployments.

Resource Profiles

Small (Development/Testing)

Use Case: Development, testing, small teams

Components:

NeuronDB: 1 replica, 2 CPU, 4Gi memory, 50Gi storage
NeuronAgent: 1 replica, 500m CPU, 512Mi memory
NeuronMCP: 1 replica, 250m CPU, 256Mi memory
NeuronDesktop: Disabled

Total Resources:

CPU: ~3 cores
Memory: ~5Gi
Storage: 50Gi

Workload Capacity:

Concurrent users: 10-50
Requests/second: 100-500
Vector operations: 1K-10K/day

Medium (Production - Small)

Use Case: Small production deployments, <100 users

Components:

NeuronDB: 1 replica, 4 CPU, 8Gi memory, 200Gi storage
NeuronAgent: 2 replicas, 1 CPU, 1Gi memory each
NeuronMCP: 1 replica, 500m CPU, 512Mi memory
NeuronDesktop: Enabled (2 replicas API, 2 replicas frontend)

Total Resources:

CPU: ~7 cores
Memory: ~11Gi
Storage: 200Gi

Workload Capacity:

Concurrent users: 50-200
Requests/second: 500-2K
Vector operations: 10K-100K/day

Large (Production - Medium)

Use Case: Medium production deployments, 100-500 users

Components:

NeuronDB: 1 replica, 8 CPU, 16Gi memory, 500Gi storage
NeuronAgent: 3 replicas, 2 CPU, 2Gi memory each
NeuronMCP: 2 replicas, 1 CPU, 1Gi memory each
NeuronDesktop: Enabled (3 replicas API, 3 replicas frontend)

Total Resources:

CPU: ~20 cores
Memory: ~26Gi
Storage: 500Gi

Workload Capacity:

Concurrent users: 200-1000
Requests/second: 2K-10K
Vector operations: 100K-1M/day

Enterprise (Production - Large)

Use Case: Large production deployments, 500+ users

Components:

NeuronDB: 1 replica, 16 CPU, 32Gi memory, 1Ti storage
NeuronAgent: 5 replicas, 4 CPU, 4Gi memory each
NeuronMCP: 3 replicas, 2 CPU, 2Gi memory each
NeuronDesktop: Enabled (5 replicas API, 5 replicas frontend)

Total Resources:

CPU: ~46 cores
Memory: ~58Gi
Storage: 1Ti

Workload Capacity:

Concurrent users: 1000+
Requests/second: 10K+
Vector operations: 1M+/day

Storage Sizing

Base Storage Requirements

OS and binaries: ~10Gi
PostgreSQL data: Variable
WAL files: ~10% of data size
Logs: ~5Gi per month
Backups: 2-3x data size (if local)

Growth Projections

Plan storage for 1 year growth:

Small: 50Gi → 100-150Gi
Medium: 200Gi → 400-600Gi
Large: 500Gi → 1-1.5Ti
Enterprise: 1Ti → 2-3Ti

Storage Class Recommendations

Development: Standard SSD
Production: Premium SSD or GP3
Enterprise: Premium SSD with IOPS optimization

Network Bandwidth

Estimated Requirements

Small: 100 Mbps
Medium: 1 Gbps
Large: 10 Gbps
Enterprise: 25 Gbps+

Factors Affecting Bandwidth

Vector embedding size
Query frequency
Replication (if enabled)
Backup operations

CPU Sizing

NeuronDB CPU

Base: 2 CPU for PostgreSQL
Vector operations: +1 CPU per 10K vectors/second
Concurrent queries: +0.5 CPU per 100 concurrent

NeuronAgent CPU

Base: 500m per replica
Request handling: +100m per 100 req/s
Background workers: +200m per worker

NeuronMCP CPU

Base: 250m per replica
MCP requests: +50m per 50 req/s

Memory Sizing

NeuronDB Memory

Base: 4Gi for PostgreSQL
Shared buffers: 25% of total memory
Work memory: 256Mi per connection
Vector cache: 1Gi per 1M vectors

NeuronAgent Memory

Base: 512Mi per replica
Request buffer: 100Mi per 100 req/s
Cache: 200Mi per replica

NeuronMCP Memory

Base: 256Mi per replica
Request buffer: 50Mi per 50 req/s

Autoscaling Recommendations

HPA Configuration

neuronagent:
  autoscaling:
    enabled: true
    minReplicas: 2
    maxReplicas: 10
    targetCPUUtilizationPercentage: 70

KEDA Configuration

neuronagent:
  autoscaling:
    keda:
      enabled: true
      minReplicas: 2
      maxReplicas: 20
      triggers:
        http:
          enabled: true
          threshold: "100"
        queueDepth:
          enabled: true
          targetValue: "100"

Node Requirements

Minimum Node Specs

Small: 4 CPU, 8Gi memory nodes
Medium: 8 CPU, 16Gi memory nodes
Large: 16 CPU, 32Gi memory nodes
Enterprise: 32+ CPU, 64Gi+ memory nodes

Node Affinity

Use node selectors for production:

neurondb:
  nodeSelector:
    node-type: database
  affinity:
    podAntiAffinity:
      preferredDuringSchedulingIgnoredDuringExecution:
        - weight: 100
          podAffinityTerm:
            labelSelector:
              matchExpressions:
                - key: app.kubernetes.io/component
                  operator: In
                  values: [neurondb]
            topologyKey: kubernetes.io/hostname

Monitoring Resource Usage

Key Metrics

CPU utilization: <70% average
Memory utilization: <80% average
Disk I/O: Monitor p95 latency
Network: Monitor bandwidth usage

Alerts

Configure alerts for:

CPU > 80% for 5 minutes
Memory > 90% for 5 minutes
Disk usage > 80%
Pod restarts > 3 in 10 minutes

Cost Optimization

Right-Sizing Tips

Start with medium profile, scale based on metrics
Use HPA for dynamic scaling
Enable resource quotas
Use spot instances for non-critical workloads
Optimize storage class based on access patterns

Resource Quotas

apiVersion: v1
kind: ResourceQuota
metadata:
  name: neurondb-quota
  namespace: neurondb
spec:
  hard:
    requests.cpu: "20"
    requests.memory: 30Gi
    limits.cpu: "40"
    limits.memory: 60Gi
    persistentvolumeclaims: "5"

Example Configurations

values-small.yaml

neurondb:
  resources:
    requests: {cpu: "2", memory: "4Gi"}
    limits: {cpu: "4", memory: "8Gi"}
  persistence:
    size: 50Gi

neuronagent:
  replicas: 1
  resources:
    requests: {cpu: "500m", memory: "512Mi"}
    limits: {cpu: "1", memory: "1Gi"}

values-production.yaml

neurondb:
  resources:
    requests: {cpu: "8", memory: "16Gi"}
    limits: {cpu: "16", memory: "32Gi"}
  persistence:
    size: 500Gi

neuronagent:
  replicas: 3
  autoscaling:
    enabled: true
    minReplicas: 3
    maxReplicas: 10
  resources:
    requests: {cpu: "2", memory: "2Gi"}
    limits: {cpu: "4", memory: "4Gi"}