Overview
Dedicated CPU Pods give you exclusive, non-shared access to physical CPU cores. Nothing else on the host can use the cores assigned to your Pod, so performance stays consistent even when neighbours are busy. This article covers what you get and when the extra cost is worth it.
What "dedicated" means
- CPU: the cores shown in your plan are reserved for your Pod alone. No noisy neighbours, predictable latency and steady throughput.
- Hyper-threading: plans are described in vCPU terms; where a host exposes SMT, a vCPU is a hardware thread on a dedicated core.
- RAM, disk and network: fully dedicated, just like other ALYPOD Pods.
Dedicated Pods are billed at a higher rate than Shared CPU Pods because the cores cannot be oversubscribed to other customers.
Ideal workloads
- Production databases (MySQL, PostgreSQL, Redis) with latency-sensitive queries.
- Game servers and real-time applications.
- Video/audio transcoding and build pipelines with long CPU-bound tasks.
- High-traffic web applications with sustained load.
Deploying and verifying
- Sign in to account.alypod.com and open the ALYPOD Console.
- Click Deploy Pod, choose a Dedicated CPU plan and an OS template.
- After boot, confirm your core count:
bash
nproc
lscpu | grep -E 'Model name|CPU\(s\)'
- Watch utilisation on the Pod's Graphs tab in the console.
Troubleshooting
- Problem: a single process is slow despite spare cores. Solution: most software is single-threaded; check whether the app supports multiple workers.
- Problem: need more cores temporarily for a migration. Solution: resize the Pod up in the console, then resize back down when finished.
Related articles
- [Shared CPU Pods explained]
- [Resizing a Pod (upgrading resources)]
- [How to deploy a CPU Pod]
