Pricing
Dashboard
Powered by Node.js 22 LTS & PM2 Cluster

Node.js Hosting built for scale, tuned for JavaScript developers

From Express APIs to NestJS microservices and Socket.IO real-time apps — we run your Node.js code on PM2-managed clusters with Nginx reverse proxy, Redis cache, and zero-downtime Git deploys. Free migration in under 24 hours.

  • PM2 cluster mode (multi-core)
  • Git push-to-deploy
  • Node 18 / 20 / 22 LTS
  • npm / yarn / pnpm
  • SSH + root-like access
  • Free Let’s Encrypt SSL
  • MongoDB / PostgreSQL / Redis
  • 24/7 Node.js experts
★★★★★ 4.9/5 from 980+ developers 30-day money-back
5× faster
SSL active
PM2 cluster
Git deploy
~/apps/my-express-api — zsh
$ git push origin main
Enumerating objects: 24, done.
Counting objects: 100% (24/24), done.
Writing objects: 100% (13/13), 2.4 KiB
remote: → SylhetHost: deploying Node.js app…
remote: ✓ npm ci --omit=dev (3.2s)
remote: ✓ pm2 reload ecosystem.config.js
remote: ✓ Nginx upstream reloaded (0 downtime)
remote: ✓ Health check 200 OK
remote: Deploy completed in 12.4s
$ pm2 status
┌─────────────┬──────┬────────┐
│ app name │ stat │ cpu │
├─────────────┼──────┼────────┤
│ my-express │ online │ 2% │
│ my-express │ online │ 3% │
│ my-express │ online │ 2% │
│ my-express │ online │ 4% │
└─────────────┴──────┴────────┘
$
zsh pm2 logs redis-cli mongo
6,800+ Node.js Apps Hosted
99.99% Uptime SLA
< 24h Free Migration
45 ms Avg Response (BD)

Why developers choose SylhetHost for Node.js

Generic shared hosting kills Node.js with single-thread limits, no PM2, no WebSockets. We built a Node-first platform where every layer — Nginx, PM2, Redis, the kernel — is tuned for the event loop.

Generic cPanel Hosting

Shared PHP stack, not built for Node

Apache + mod_php doesn’t speak Node. You get 1 process, no cluster, killed sockets, and Phusion Passenger hacks that crash on real traffic.

  • × No PM2 / no cluster mode
  • × No persistent WebSockets
  • × No native npm / yarn
  • × No Redis / no MongoDB
  • × No Git push-to-deploy

DIY VPS / Cloud

You run your own server

Spin up DigitalOcean / AWS, install Node, configure Nginx, set up PM2, harden the firewall, handle SSL renewals, monitor at 3 AM. Hope nothing breaks while you sleep.

  • × You patch + secure the OS
  • × No Laravel / BD-support team
  • × Per-hour cloud billing surprises
  • × No free migration help
  • × You monitor 24/7 yourself

PM2 cluster mode = N× the throughput

Node.js is single-threaded by default — it uses one CPU core. PM2 cluster mode spawns one worker per core, so a 4-core VPS handles 4× the requests. Below: requests/sec on a hello-world Express endpoint (wrk, 100 connections, 30s).

  • Multi-core utilization out of the box
  • Zero-downtime reloads via PM2 cluster mode
  • Nginx reverse proxy with HTTP/2 + keep-alive
  • Redis-backed session + response cache
See Node.js Plans

Requests / second (Express hello-world)

Node single process ~1,200 rps
PM2 cluster (2 cores) ~2,300 rps
PM2 cluster (4 cores) ~4,500 rps
PM2 + Redis cache ~6,600 rps

Node.js hosting plans for every workload

From a single Express API to a fleet of NestJS microservices. All plans include PM2, Git deploys, SSH, free SSL, daily backups, and 24/7 expert support. Switch any time, no lock-in.

Save 20% — 2 months free

Starter

Side projects & small APIs

500 /mo

 

Get Started
  • 1 Node.js app
  • 1 GB RAM / 1 vCPU
  • PM2 cluster, 2 workers
  • 10 GB NVMe storage
  • 100 GB bandwidth
  • Free SSL + daily backups
  • 1 MongoDB database (256 MB)
  • SSH + Git deploy

Pro

High-traffic apps & microservices

2,500 /mo

 

Scale Up
  • 15 Node.js apps
  • 8 GB RAM / 4 vCPU
  • PM2 cluster, 8 workers
  • 100 GB NVMe storage
  • 1 TB bandwidth
  • Free SSL + hourly backups
  • Unlimited DBs (Mongo/PG/MySQL)
  • Redis 1 GB + BullMQ workers
  • Multi-region replica (read)
  • Dedicated IP + WAF

Enterprise

Mission-critical at scale

5,500 /mo

 

Talk to Sales
  • Unlimited apps
  • 16 GB RAM / 8 vCPU
  • PM2 cluster + auto-scale
  • 250 GB NVMe storage
  • 5 TB bandwidth
  • Free SSL + hourly backups
  • Dedicated MongoDB replica set
  • Redis cluster + SQS queues
  • Multi-AZ HA + DR failover
  • Dedicated Node.js SRE
  • 99.99% SLA + 1h response

6 layers every Node.js request passes through

From the browser to your Express handler and back — each layer is tuned, cached, and observed so you can ship features instead of debugging infrastructure.

Cloudflare CDN

Static assets cached at 280+ edge POPs

WAF + DDoS

OWASP rules + L3/L4/L7 mitigation

Nginx Reverse Proxy

HTTP/2, SSL termination, keep-alive

PM2 Cluster

Multi-core workers, zero-downtime reload

Redis Cache

Sessions, rate-limit, response cache

Mongo / Postgres

Connection pool, indexed queries

Deploy with git push — zero downtime, every time

Connect your GitHub / GitLab / Bitbucket repo. Every push to main triggers a build: npm ci, run migrations, spin up new PM2 workers, drain the old ones, health-check, done. Rollback in one click if anything goes sideways.

  • 1
    Push to main

    git push origin main — the webhook fires in under 2 seconds.

  • 2
    Install + build

    npm ci / yarn install / pnpm install, then npm run build if present.

  • 3
    PM2 reload

    New workers boot, old ones drain in-flight requests, zero dropped connections.

  • 4
    Health check + live

    GET /health returns 200 → traffic routed. Fail → instant automatic rollback.

Start Deploying
deploy.log — my-express-api
$ git push origin main Enumerating objects: 24, done. Writing objects: 100% (13/13), 2.4 KiB remote: → SylhetHost build pipeline started remote: ✓ Cloning repository (1.1s) remote: ✓ npm ci --omit=dev (3.2s) remote: ✓ Running migrations (0.8s) remote: ✓ Building TypeScript (2.4s) remote: → Reloading PM2 ecosystem… remote: ✓ PM2 workers reloaded (0 downtime) remote: ✓ Nginx upstream reloaded remote: ✓ Health check 200 OK /health remote: ✓ Old workers drained (0 dropped) $ Deploy completed in 12.4s ✨

Everything you need to ship Node.js, nothing you don’t

Real SSH access, real npm, real PM2, real Node REPL. Not a sandboxed cPanel plugin — the actual tools, on the actual server, the way you’d set them up yourself.

SSH + SFTP Access

Full SSH shell with bash/zsh. Connect via key auth, run any command, edit files in place.

ssh user@app.sylhethost.com

npm / yarn / pnpm

All three package managers pre-installed. Lockfiles respected. Private registries supported.

pnpm install --frozen-lockfile

node CLI + REPL

Switch Node 18 / 20 / 22 per app via nvm. Drop into the REPL to inspect runtime.

node --eval "require('./db')"

PM2 Process Manager

Cluster mode, reload, logs, metrics. Pre-wired ecosystem.config.js on every app.

pm2 reload ecosystem.config.js

Git Push-to-Deploy

GitHub / GitLab / Bitbucket webhooks. Auto-build, auto-migrate, auto-health-check.

git push origin main

Clinic.js Profiler

CPU flame graphs + memory leaks + event-loop blocking. Built into every plan.

clinic doctor -- node app.js

WebSockets / Socket.IO

Persistent WS connections, Nginx upgrade support, sticky sessions, Redis adapter.

io.adapter(redisAdapter(uri))

Deploy Hooks

Pre-deploy, post-deploy, on-failure hooks. Slack / Discord / email notifications.

scripts.post-deploy: "npm run cache:clear"

The right database for the right job — all included

Node.js apps speak many databases. We pre-wire MongoDB for documents, PostgreSQL for relations, MySQL for legacy, and Redis for everything fast. Connection pooling, replicas, backups — all handled.

MongoDB

v7.0 · Wire tiger

Document database for Mongoose / native driver. Replica set on Pro+, change streams enabled.

  • Indexes, aggregations, transactions
  • Connection pooling (max 100)
  • mongosh + Compass pre-installed
  • Daily snapshots, 30-day retention

PostgreSQL

v16 · JSONB + pgvector

The most feature-complete open-source SQL DB. Prisma, TypeORM, Knex, Drizzle all supported.

  • JSONB, full-text search, CTEs
  • pgvector for AI embeddings
  • Read replicas on Pro+
  • pg_partman for time-series

Redis

v7.2 · AOF + RDB

Cache, sessions, rate-limit, queues. Bull / BullMQ backed, Socket.IO adapter ready.

  • Streams, pub/sub, Lua scripts
  • Persistence AOF + RDB
  • Redis Sentinel HA on Pro+
  • 1 GB → 16 GB scalable

MySQL

v8.0 · InnoDB

Legacy relational DB for Sequelize, TypeORM MySQL driver, or migrations from PHP stacks.

  • CTEs, window functions, JSON
  • Read replicas on Pro+
  • phpMyAdmin optional
  • Percona Toolkit installed

Meilisearch

v1.6 · typo-tolerant

Sub-50ms full-text search, drop-in Algolia alternative. One-click enable, instant reindex.

  • Typo tolerance, facets, filters
  • Geo search, custom ranking
  • Auto-sync via webhooks
  • Multi-tenant API keys

BullMQ / SQS

v5 · Redis-backed

Background jobs, scheduled tasks, retries, rate-limiting. SQS compatible on Enterprise.

  • Priority queues, delayed jobs
  • Supervisor-managed workers
  • Auto-restart on deploy
  • Dashboard at /horizon-style UI

8 layers of Node.js-aware security

Security built for the runtime — from .env protection to WAF rules tuned for Express routing. We patch, you ship.

  • Web Application Firewall — OWASP + Express-specific rules
  • Free Auto-Renew SSL — Let’s Encrypt + HSTS preload
  • Imunify360 + Malware Scan — real-time node_modules audit
  • DDoS Mitigation — L3/L4/L7, 2 Tbps+ capacity
  • SSH Key Auth + 2FA — no password login on SSH/sFTP
  • Isolated PM2 per app — one app can’t read another’s .env
  • Daily Off-site Backups — code + DB + Redis, 30-day retention
  • 24/7 Threat Monitoring — SOC team + auto-incident response

First-class support for every Node.js framework

Pre-configured, pre-tuned, pre-wired. Just npm install and go — the infrastructure side is already done.

Express

Pre-wired Nginx reverse proxy, body-parser, helmet, compression. Health-check endpoint ready.

NestJS

DI, modules, controllers, providers. Swagger UI auto-mounted at /api/docs.

Socket.IO

WebSocket server with Redis adapter, sticky sessions, multi-node broadcast.

Fastify

High-performance alternative to Express. JSON schema validation included.

Prisma

Type-safe ORM with migrate, studio, introspect. Postgres + MySQL + MongoDB.

Mongoose

MongoDB ODM with schemas, validators, middleware, population.

BullMQ

Redis-backed job queue with priorities, retries, scheduled jobs, dashboard.

Passport.js

OAuth/JWT/local strategies pre-wired. Google, GitHub, Facebook callbacks.

Nodemailer

SMTP + Mailgun + Postmark + SES. Per-app mail logs, retries, webhooks.

Stripe SDK

Webhooks reachable out of the box. Signature verification configured.

Next.js

SSR / SSG / API routes. Standalone build mode, image optimization on.

S3 / Cloud Storage

Pre-wired AWS SDK + MinIO-compatible local. Multer + sharp ready.

Pino + OpenTelemetry

Structured JSON logs, distributed tracing, OTLP exporter pre-configured.

JWT / jsonwebtoken

RS256 / HS256, refresh tokens, JWKS endpoint cached for fast verification.

Unleash / PostHog

Feature flags backed by Redis. A/B testing, gradual rollouts.

GitHub Actions CI

Auto-deploy on green builds. PR preview environments on Pro+.

Versions you can pick, not versions forced on you

Per-app Node.js selector means legacy Node 18 apps coexist with brand-new Node 22 apps. Switch in one click.

22 Node.js 22
20 Node.js 20 LTS
18 Node.js 18 LTS
10 npm 10
9 pnpm 9
4 yarn 4
5 PM2 5
7 MongoDB 7
16 PostgreSQL 16
8 MySQL 8
7 Redis 7
1.25 Nginx 1.25

Migrate from any host in under 24 hours

Coming from Heroku, Render, Railway, Vercel, DigitalOcean, AWS, or your own VPS? Our migration team handles the move end-to-end — code, databases, DNS, SSL — while you sleep. Zero downtime, zero data loss.

1

Share Access

Tell us your current host + repo. We sign an NDA, you grant read-only access.

2

Replicate + Test

We clone your app, run npm ci, migrate the database, run smoke tests on staging URL.

3

DNS Cutover

Update A record / CNAME. SSL auto-issued. Old host stays live as fallback for 7 days.

4

Verify + Optimize

Run load tests, tune PM2 workers + Redis, profile slow queries. Hand you the keys.

Bangladeshi startups running on SylhetHost Node.js

From fintech to e-commerce to real-time education platforms — here’s what teams shipped after migrating to us.

FastrPay
fintech · Dhaka

“Cut our API latency from 280 ms to 45 ms after migrating from Heroku. PM2 cluster mode + Redis cache is a cheat code.”

Faster API
0 Downtime
62% Cost saved
ShopList BD
e-commerce · Chittagong

“Black Friday 2024: 18,000 concurrent Socket.IO connections, not a single dropped cart. PM2 + Redis adapter handled it without breaking a sweat.”

18k Concurrent
99.99% Uptime
2.4s Avg load
EduConnect
edtech · Sylhet

“Migrated 14 NestJS microservices from AWS in 18 hours. The migration team did everything — we just kept coding the next feature.”

14 Services
18h Migration
Throughput

Node.js hosting — questions, answered

Everything you wanted to know about running Node.js on SylhetHost. Don’t see your question? Live chat with a real Node.js engineer, 24/7.

Can I run Socket.IO / WebSockets on every plan?
Yes. Every plan — Starter through Enterprise — supports persistent WebSocket connections out of the box. Nginx is pre-configured with the Upgrade/Connection headers needed for ws:// and wss://, sticky sessions are enabled by default for multi-worker PM2, and on Business+ we include the Socket.IO Redis adapter so you can broadcast across multiple workers without message loss. No proxy rewriting, no port forwarding, no extra cost.
How does PM2 cluster mode work and why does it matter?
Node.js is single-threaded by default — one process, one CPU core, one event loop. On a 4-core server you’d be wasting 75% of your hardware. PM2 cluster mode spawns one Node.js worker per CPU core, each running its own event loop, and load-balances incoming requests across them. The result: roughly N× throughput on N cores, automatic worker restart on crash, and zero-downtime reloads (pm2 reload) that boot new workers before draining the old ones. We pre-wire ecosystem.config.js on every app so it just works.
Can I pick which Node.js version my app runs on?
Yes — per-app. We use nvm under the hood, so each app can pin its own Node version. Supported versions: Node 18 LTS, Node 20 LTS, and Node 22 (current). Switching takes one click in the control panel or one line over SSH (nvm use 22). This means a legacy Node 18 app and a brand-new Node 22 app can live on the same VPS without conflicts.
Do you support npm, yarn, and pnpm?
All three. npm 10, yarn 4, and pnpm 9 are pre-installed globally. The deploy pipeline auto-detects which package manager your repo uses based on the lockfile (package-lock.json → npm, yarn.lock → yarn, pnpm-lock.yaml → pnpm) and runs the correct install command with --frozen-lockfile / --omit=dev for production. Private npm registries are supported via .npmrc with auth tokens stored in an encrypted vault.
How does the Git push-to-deploy pipeline work?
Connect your GitHub / GitLab / Bitbucket repo once. Every git push origin main triggers a webhook that runs: npm ci, npm run build (if present), database migrations (if detected), pm2 reload ecosystem.config.js, Nginx upstream reload, and a GET /health smoke test. Total time: typically 8–15 seconds. If the health check fails, the deploy auto-rollbacks to the previous release — no action needed on your end. Roll back manually any time via the control panel.
Do I get real SSH access?
Yes — full SSH shell with bash or zsh, key-based auth (password login disabled by default), and SFTP for file transfers. You can run npm, node, pm2, mongosh, psql, redis-cli, git, and any standard Linux utility. Root access is available on Enterprise. We never jailbreak or chroot you into a fake shell — this is the real thing.
What databases are included?
Every plan includes MongoDB 7, PostgreSQL 16, MySQL 8, and Redis 7. The number of databases and storage limits vary by plan (see the pricing table above). Meilisearch (full-text search) and BullMQ (job queues) are one-click add-ons. On Pro+, you get read replicas for Postgres and MongoDB. On Enterprise, you get dedicated replica sets and Redis Sentinel for high availability.
Is there a money-back guarantee?
Yes — 30 days, no questions asked. If SylhetHost Node.js isn’t right for you, email support@sylhethost.com within 30 days of signup and we’ll refund 100% of your hosting fee. Domain registrations and SSL certificate purchases are non-refundable per ICANN policy.
Do you help with migration from Heroku / Render / Vercel / AWS?
Yes — free on every plan. Our migration team handles code, databases, environment variables, DNS cutover, SSL re-issue, and post-migration load testing. Typical turnaround is under 24 hours for a single app, under 72 hours for a multi-service architecture. We sign an NDA before you share any access. Zero-downtime cutover is the default — your old host stays live for 7 days as a fallback.

Ship your Node.js app today — free for 30 days

Spin up your first Node.js app in under 5 minutes. No credit card required for the trial. If we don’t earn your business in 30 days, walk away — no charge, no questions.

30-day money-back Free migration < 24h 24/7 Node.js experts No credit card to start