Cloud VPS
For websites, APIs, development environments and smaller databases. Budget memory for your application, database and cache, then size for observed peaks.
Compare VPS plansPOWOKE / GUIDES
Server buying and deployment guide
Choose infrastructure around your workload, audience and operating requirements. Compare VPS, dedicated servers, GPU hosting and virtualization hosts, then plan a tested deployment.
Match the hosting model to your workload
Dedicated servers
For sustained database workloads, game servers and applications requiring exclusive hardware. Compare CPU generation, ECC memory, disk layout and remote management.
Explore dedicated serversGPU servers
For AI training, LLM inference and rendering. Size VRAM around model precision and workload, then check GPU count, drivers, CUDA and framework compatibility.
Compare GPU optionsVirtualization hosts
For VPS platforms and private clouds. Plan CPU allocation, memory, storage I/O, IP requirements and hypervisor licensing around the number of virtual machines.
Explore virtualization hostsEvaluate location, resources and total cost
Audience and network paths
Consider customer networks alongside databases, payment services and external APIs. Test latency and packet loss during peak hours from relevant ISPs; distance alone does not predict response time.
CPU and memory
Measure baseline and peak application, database and cache usage. Leave capacity for updates, backups and growth. A vCPU does not imply a dedicated physical core; check allocation policies.
Storage and recovery
Separate capacity, random I/O and sustained throughput requirements. Check disk expansion, backup retention and restore procedures. RAID and snapshots do not replace an independent backup.
Transfer and operating systems
Compare port speed separately from monthly transfer allowances, overage rates and throttling policies. Check supported OS versions, application compatibility and Windows or other licensing costs.
Confirm these details before checkout
- The selected location and exact CPU, RAM, storage and GPU configuration.
- The billing-term total, renewal price, currency and promotion conditions.
- Transfer allowances, IP allocations, software licenses and optional service charges.
- Stock, expected provisioning time and any installation or migration arrangements.
- Administrator access, support scope, SLA measurement and applicable service terms.
- Refund eligibility, backup responsibilities and maintenance required for upgrades.
Deploy securely and maintain your server
These examples target Debian / Ubuntu Linux. Check your distribution, SSH port and access permissions first. Use the appropriate tools for Windows or custom images, and prepare a restorable backup before changes.
First login and access controls
Update the OS, create a non-root administrator and install your SSH public key. Verify key access in a second session before disabling password login. Allow your actual SSH port before enabling a firewall, and retain console access and a restorable backup.
sudo apt update
sudo apt upgrade
sudo adduser deploy
sudo usermod -aG sudo deploy
sudo ufw allow OpenSSH
sudo ufw status verboseDeploy a Docker workload
Install Engine and the Compose plugin using Docker's distribution-specific instructions. Pin image versions or digests; never expose the Docker socket or publish database administration ports on 0.0.0.0. Validate a staging deployment before switching traffic.
docker version
docker compose version
docker compose config --quiet
docker compose up -d
docker compose ps
docker compose logs --tail=100Separate routing and application faults with MTR
Replace YOUR_SERVER_IP with an instance you are authorized to test. Measure the outbound path from the client and the return path from the server. Intermediate ICMP loss can indicate rate limiting; correlate loss continuing to the destination with application errors.
sudo apt install mtr-tiny
mtr -rwzc 50 YOUR_SERVER_IP
mtr --tcp --port 443 -rwzc 50 YOUR_SERVER_IP
curl -o /dev/null -sS -w 'dns=%{time_namelookup} connect=%{time_connect} ttfb=%{time_starttransfer} total=%{time_total}\n' https://YOUR_DOMAIN/Backups, upgrades and support tickets
Create a restorable off-site backup before changes, rehearse recovery and monitor disk capacity and certificate expiry. Network tickets should include timestamps and timezone, source ISP, destination IP, protocol, port and a redacted MTR. Never include passwords, tokens or customer personal data.
df -h
free -h
systemctl --failed
journalctl -p err --since '1 hour ago'Validate before switching production traffic
Test the application, database and scheduled jobs using a temporary hostname. Verify DNS, HTTPS, access controls and alerts. For migrations, synchronize data and prepare rollback steps before cutover; retain the old environment until validation is complete.
Design a multi-tenant Proxmox hosting platform
Plan tenant isolation, IP allocation and recovery before choosing a virtualization host. This is an architecture checklist, not a benchmark or a promise that a specific IP block is included.
Read the architecture guide →Plan a GPU server around your AI workload
Define memory, isolation, storage and serving requirements for image generation and LLM inference before comparing GPU configurations.
Discuss GPU requirements →Need help sizing a deployment?
Share user locations, workload details, resource usage, budget and launch date so support can confirm suitable configurations and provisioning arrangements.