|
| 1 | +--- |
| 2 | +meta: |
| 3 | + title: Choosing bewteen shared or dedicated CPUs |
| 4 | + description: Find out how to choose on shared or dedicated CPUs depending on your workloads. |
| 5 | +content: |
| 6 | + h1: Choosing bewteen shared or dedicated CPUs |
| 7 | + paragraph: Find out how to choose on shared or dedicated CPUs depending on your workloads. |
| 8 | +categories: |
| 9 | + - compute |
| 10 | +dates: |
| 11 | + validation: 2025-04-18 |
| 12 | + posted: 2025-04-18 |
| 13 | +tags: instance, type, stardust, range, vcpu, hyperthread, core, ram, bandwidth, dedicated, shared, memory, hypervisor, vm, storage, dev1, play2, gp1, pro2, ent1, gpu, arm, learning, development, production, production-optimized, cost-optimized, memory-optimized, storage-optimized |
| 14 | +--- |
| 15 | + |
| 16 | +When deploying a Scaleway Instance, selecting the appropriate CPU provisioning is key for optimizing performance and cost. |
| 17 | + |
| 18 | +Two types of CPU provisioning are available for Instances: **shared** and **dedicated** CPUs. |
| 19 | + |
| 20 | +## Comparison of shared and dedicated CPU Instances |
| 21 | + |
| 22 | +| Feature | Shared CPU | Dedicated CPU | |
| 23 | +|---------------------------------|----------------------------------------------------------|----------------------------------------------------------------| |
| 24 | +| Performance consistency | Variable – depends on other workloads on the host | High – consistent and predictable performance | |
| 25 | +| Cost | Lower | Higher | |
| 26 | +| Use case | Dev/test environments, low-traffic apps | Production apps, high-performance workloads | |
| 27 | +| CPU access | Shared with other instances | Exclusive access to physical cores | |
| 28 | +| Best for | Personal blogs or forums, staging environments | CI/CD, eCommerce, gaming, ML workloads | |
| 29 | +| Resource contention risk | Possible during peak usage | None | |
| 30 | +| Latency sensitivity | Not suitable for latency-sensitive apps | Ideal for latency-critical applications | |
| 31 | + |
| 32 | +## Shared CPU Instances |
| 33 | +Shared CPU Instances, including [Learning](/instances/reference-content/learning/) and [Cost-Optimized](/instances/reference-content/cost-optimized/), are cost-effective virtual machines in which CPU resources are shared among multiple Instances. |
| 34 | +This means multiple virtual CPU cores are allocated to these Instances, but the physical CPU cores available on the hypervisors' hardware are shared among them. |
| 35 | +As a result, Instances share physical CPU time, and during peak demand from other Instances on the same host, your workloads might temporarily slow down due to CPU contention (also known as "CPU steal"). |
| 36 | + |
| 37 | +### Typical use cases include |
| 38 | +- Development and staging environments |
| 39 | +- Low to medium-traffic websites |
| 40 | +- Personal blogs and forums |
| 41 | +- Applications tolerant to occasional performance variability |
| 42 | +- Worker nodes in container orchestration clusters such as Kubernetes |
| 43 | +- Experimental or proof-of-concept projects |
| 44 | +- Small-scale applications with limited traffic |
| 45 | + |
| 46 | +### Summary |
| 47 | +- Shared CPU Instances provide an affordable solution for non-critical workloads. |
| 48 | +- CPU performance is less predictable and may fluctuate depending on neighboring workloads ("noisy neighbours"). |
| 49 | +- During peak usage, your workloads might experience temporary slowdowns due to CPU steal. |
| 50 | + |
| 51 | +## Dedicated CPU Instances |
| 52 | +Dedicated CPU Instances, including [Production-Optimized](/instances/reference-content/production-optimized/) and [Workload-Optimized](/instances/reference-content/workload-optimized/), provide exclusive access to physical CPU cores. |
| 53 | +This ensures consistent and predictable performance at all times. Dedicated CPU Instances are perfect for applications that require high CPU utilization and low latency. |
| 54 | + |
| 55 | +### Typical use cases include |
| 56 | +- Production applications with high CPU demands |
| 57 | +- eCommerce platforms and business-critical services |
| 58 | +- Game servers requiring low latency |
| 59 | +- CI/CD pipelines |
| 60 | +- Audio and video transcoding |
| 61 | +- Machine learning and scientific computing |
| 62 | +- Real-time data processing and analytics |
| 63 | +- High-traffic websites and applications |
| 64 | + |
| 65 | +### Summary |
| 66 | +- Dedicated CPU allocation ensures consistent and predictable performance. |
| 67 | +- No risk of performance degradation due to neighboring workloads. |
| 68 | +- Dedicated CPU Instances are more expensive than shared CPU Instances, but offer guaranteed CPU performance. |
| 69 | + |
| 70 | +## Choosing the best configuration |
| 71 | + |
| 72 | +Choose **shared CPU** Instances if: |
| 73 | + |
| 74 | +- You're running non-critical or experimental workloads |
| 75 | +- Budget is a priority over performance consistency |
| 76 | + |
| 77 | +Choose **dedicated CPU** Instances if: |
| 78 | + |
| 79 | +- Your application requires stable, predictable CPU performance |
| 80 | +- You are in a production environment with strict performance requirements |
| 81 | + |
| 82 | +Consider your needs and workload requirements to choose the best CPU provisioning option for your Scaleway Instance. |
| 83 | + |
| 84 | +For more details about available instance types, refer to [Choosing the best Scaleway Instance type for your workload](/instances/reference-content/choosing-instance-type/). |
0 commit comments