Single-process SQLite n8n deployments stall whenever bursty CRM webhooks collide with multi-megabyte JSON Code nodes or AI chains. On BotFoundry, we size EXECUTIONS_MODE=queue topologies with dedicated webhook listeners, Redis AOF brokers, and PostgreSQL 16 workers on 8 GB to 32 GB DDR5 KVM slices.
Public rates retrieved 2026-11-12 from official vendor pricing pages cited below; promotional & renewal prices may change — always verify exact specs, term length, and renewal rates on the vendor’s official checkout page.
| Concurrency Scale | vCPU | DDR5 RAM | OS / Runtime | /dev/shm & Watchdog Guard | Public Tier Match (2026-11-12) |
|---|---|---|---|---|---|
| Starter Queue (2 Workers) | 2 vCPU KVM | 8 GB DDR5 | Ubuntu 24.04 1-Click n8n | Redis AOF + Prune 168h |
Hostinger KVM 2 ($5.99/mo) |
| AI Chain & CRM Burst (6 Workers) | 4 vCPU KVM | 16 GB DDR5 | Ubuntu 24.04 Docker | MaxOld=2048 + Postgres 16 |
Hostinger KVM 4 ($10.49/mo) |
| Agency Multi-Tenant (15 Workers) | 8 vCPU KVM | 32 GB DDR5 | Ubuntu 24.04 NVMe | 3x Webhook + PgBouncer |
Hostinger KVM 8 ($19.99/mo) |
| High-Throughput Bus (30+ Workers) | 8–12 Dedicated VDS | 64 GB ECC | Ubuntu 24.04 Dedicated | Dedicated Cores + WAL Disk |
UltaHost VDS 64GB ($98/mo) |
x-n8n-env: &n8n-env
DB_TYPE: postgresdb
EXECUTIONS_MODE: queue
QUEUE_BULL_REDIS_HOST: redis
OFFLOAD_MANUAL_EXECUTIONS_TO_WORKERS: "true"
EXECUTIONS_DATA_PRUNE: "true"
EXECUTIONS_DATA_MAX_AGE: 168
services:
n8n-main:
image: docker.n8n.io/n8nio/n8n:latest
restart: unless-stopped
environment: *n8n-env
n8n-webhook:
image: docker.n8n.io/n8nio/n8n:latest
restart: unless-stopped
command: webhook
environment: *n8n-env
n8n-worker:
image: docker.n8n.io/n8nio/n8n:latest
restart: unless-stopped
command: worker --concurrency=10
deploy:
replicas: 3
resources:
limits:
memory: 1536M
environment: *n8n-envStripe, Shopify, and Telegram gateways retry or drop webhooks if your endpoint takes longer than 3–5 seconds to acknowledge. Running dedicated n8n webhook containers enqueues jobs into Redis immediately while background worker replicas process LLM prompts and file buffers asynchronously.
Saving full JSON execution snapshots across 50,000 daily runs bloats execution_entity by 15–25 GB per week. Set EXECUTIONS_DATA_SAVE_ON_SUCCESS=none on high-frequency workflows and prune completed records older than 168 hours (7 days).
Adding Qdrant/pgvector embeddings or a headless Chromium container beside n8n adds 2.5–4.0 GB of steady resident memory — making the jump from 8 GB (KVM 2) to 16 GB (Hostinger KVM 4) essential.
We allocate 1.0 to 1.5 GB RAM per n8n worker container (--concurrency=10) with explicit Docker memory: 1536M caps so a single runaway JavaScript Code node cannot crash PostgreSQL.
Without Append-Only File (AOF) persistence, an unplanned container restart wipes every queued webhook payload waiting in BullMQ memory.
Yes — Hostinger offers official 1-Click n8n and Queue Mode OS templates on KVM 1/2/4/8 plans; verify current promotional rates on their official checkout page.
BotFoundry 4-layer RAM equation, /portable chart trimming, and Windows AutoLogon watchdog rules for 1 to 50+ MetaTrader 4/5 Expert Advisor terminals.
BotFoundry sizing formulas for Ubuntu 24.04 Freqtrade Docker bots, cgroup MemoryMax bounds, chronyd NTP sync, and isolated FreqAI LightGBM retraining.
BotFoundry memory math for headless Chromium BrowserContexts, Docker --shm-size=2g shared memory pools, tini zombie reaping, and flat-bandwidth KVM nodes.
BotFoundry sizing formulas for Hummingbot V2 Controllers, single-thread 3.5GHz+ Cython loops, EMQX MQTT brokers, and TypeScript DEX Gateway containers.