Product Description Cloud Servers
A cloud server is a virtual server of a fixed or arbitrary configuration that runs on Servercore physical resources. Servercore uses the KVM hypervisor as the virtualization system for its physical resources.
Terms of use for the Cloud Servers product are defined by the document User agreement and terms of use.
In addition to standard cloud servers, Servercore allows you to create preemptible servers with a discount of up to 75%. A preemptible cloud server runs continuously for no more than 24 hours and can be stopped by Servercore at any time. Terms of use do not apply to preemptible servers.
Cloud servers can be created in different locations. For more information, see the instructions Cloud server availability by location.
Cloud servers support user types and roles, projects, and project limits and quotas.
Use cases
- hosting web projects: web applications, online stores, corporate websites, portals, and other internet resources;
- hosting software and corporate infrastructure services (CRM, ERP, databases, mail servers, and others);
- creation, development, and testing of applications;
- running containerized applications and microservice architectures;
- performing load testing and organizing CI/CD processes;
- scaling fault-tolerant web services during peak loads;
- rendering graphics, animations, and visual effects;
- performing GPU-accelerated computing for scientific and engineering tasks;
- inference of machine learning and artificial intelligence models;
- running and maintaining large language models (LLMs), chatbots, RAG systems, and AI agents;
- organizing remote workplaces and virtual desktops.
How it works
To get started, create a cloud server. You can create a server:
- in the Control Panel;
- using the OpenStack CLI;
- using Terraform.
When creating a server, you select a line and a specific configuration that fits your tasks. Two types of configurations are available for most lines:
- fixed configuration — a configuration with a fixed ratio of CPU cores, memory, and disks;
- custom configuration — a ratio of CPU cores, memory, and disks adjustable to your needs.
Depending on the tasks your cloud server will handle, you can:
- configure a server with dedicated cores;
- choose a local or network drive;
- add several additional network volumes;
- configure a network for the server;
- add security groups;
- configure access to the cloud server;
- add the cloud server to a placement group;
- or configure automation on the first OS boot using User data.
How to work with cloud servers
Once you have created a server, you can manage it:
- in the Control Panel;
- OpenStack CLI;
- or Terraform.
To access the cloud server remotely, you can connect in several ways.
You can track cloud server metrics using Metrics.
Records of operations with cloud servers are saved in audit logs.
Areas of responsibility
Areas of responsibility at Servercore correspond to the division of responsibilities in the IaaS cloud service delivery model.
Servercore provides
- a virtualization environment and computing systems;
- physical security and fault tolerance of servers, and power supply;
- collection and analysis of hypervisor and infrastructure component events;
- protection against DDoS and network attacks;
- traffic encryption and filtering.
Servercore is not responsible for
- configuration of cloud servers: the operating system, network, and firewall;
- security of the virtual network and user accounts;
- resource access control;
- cloud server backup;
- encryption and data integrity verification.
You manage the cloud server yourself: you install and configure the necessary software.
Pricing
Cloud servers are billed on a pay-as-you-go model. Funds are deducted from the balance every hour for the previous hour of using cloud server resources. Outbound traffic over 3 TB across all Cloud Platform projects is also billed. For more information, see the Billing models and prices.
What is included in the price
The price of a cloud server includes:
- 3 TB of outbound traffic per month across all Cloud Platform projects in one account;
- unlimited traffic between Servercore products;
- a library of pre-built images and applications;
- server management via API and Terraform;
- DDoS protection at L3–L4;
- cloud firewall;
- security groups.
Limits and restrictions
The total bandwidth for UDP traffic on ports 0, 16, 19, 53, 123, 1900 is limited to 500 Mbit/s.
If the threshold of 150 000 packets per second is exceeded, Servercore limits the bandwidth to 100 Mbit/s.
Servercore sets account and project limits, as well as resource quotas. For more information, see the instructions Quotas and limits
In Servercore, some TCP/UDP ports are blocked by default, including ports 25, 465, and 587. You can submit a request to unblock ports. To send emails, you can use the Servercore email service.
Access to public subnets of the e-Government (Gosuslugi portal and related services) is also blocked from the Servercore infrastructure. To unblock it, submit a request for access to e-Government subnets.
Cloud servers do not support nested virtualization. If nested virtualization is required for your tasks, consider dedicated servers or Managed Kubernetes clusters on dedicated servers.