Cloud server configurations
When creating a cloud server, you can select the number of vCPUs, RAM, local disk size (optional), and add GPUs.
Two types of configurations are available:
- fixed configurations — several ranges with different specifications, in which the resource ratio is fixed;
- custom configurations in which any ratio of resources can be specified.
Once the cloud server is created, you can change its configuration.
Processors
The fixed and custom configurations differentiate between the available processors.
Standard Line
CPU Line
Memory Line
GPU Line
Shared Line
HighFreq Line
Arbitrary configurations
Processors with DDR4 REG 2133-2933 MHz.
Processors with DDR4 REG 2133-2933 MHz.
Processors with DDR4 REG 2133-2933 MHz.
Processors with DDR4 REG 2133-2933 MHz.
Processors with DDR4 REG 2133-2933 MHz.
6354 processors with DDR4 REG 3200MHz and 9474F processors with DDR5 REG 4800MHz.
* With a cloud server load of 100%, the processor runs with Turbo Boost technology and a maximum frequency of 3.60 GHz. Since the processor is emulated, only the 3.00 GHz frequency will be displayed when testing inside the cloud server.
Processors with DDR4 REG 2133-2933 MHz.
Fixed configurations
To see the availability of configurations in the regions, see the availability matrix Cloud servers.
The fixed configuration can be selected whencloud server creation in the control panel. If you create a cloud server through the OpenStack CLI and Terraform, use the following table List of fixed-configuration flavorings in all pools.
Depending on the lineup, up to 36 vCPUs, 320 GB of RAM and 1.25 TB of local disk are available in fixed configurations.
Standard Line
A line of fixed cloud server configurations with balanced 1:4 vCPU:RAM resource allocation (except for entry-level and maximum configurations).
Suitable for most tasks, such as executing code for web services and applications, hosting online stores, and creating test environments.
Available from 1 to 36 vCPUs, 1 GB to 128 GB RAM, 8 GB to 1.25 TB local disk.
CPU Line
A line of fixed configurations in which vCPU:RAM resources are balanced in a 1:2 ratio.
Suitable for video transcoding, machine learning, data processing, building CI/CD systems and other tasks that require performance and compute speed on vCPUs.
Available from 4 to 24 vCPUs, 8 GB to 48 GB RAM, 128 GB to 512 GB local disk.
Memory Line
A line of fixed configurations in which vCPU:RAM resources are balanced at a 1:8 ratio.
The configurations provide high performance for workloads handling large data packages to accommodate demanding databases or SAP and 1C enterprise applications.
Available from 2 to 16 vCPUs, 16 GB to 128 GB RAM, 64 GB to 512 GB local disk.
GPU Line
Fixed cloud server configurations with dedicated GPUs.
Optimized for GPU computing, such as video transcoding, training neural networks, or creating remote workstations.
Available from 4 to 32 vCPUs, 1 to 4 GPUs, 32 GB to 320 GB RAM.
The GPU Line can be used with a local or network boot disk. For cloud servers, only NVIDIA® A100 40Gb, NVIDIA® A100 80Gb, or NVIDIA® A30 in the ru-7a and ru-7b pool segment can be used with a local disk.
Read more in the instructions Create a cloud server with GPU.
Shared Line
Fixed cloud server configurations with the ability to use and pay for only a portion of the vCPU.
Shared Line configurations are less expensive than other lineups and are suitable for tasks that do not require constant and full utilization of the virtual core, such as running staging, a website, bringing up a network, or training.
In Shared Line, one virtual core can be used by several clients at once. When creating such a configuration, you need to specify the share of vCPU that will be reserved for your server: 10%, 20% or 50%. Cloud server performance will never go below the specified share and may temporarily go up to 100% if other clients are not maximizing resources or a portion of the virtual core is not leased.
Use only with a network boot disk.
Available from 1 to 4 vCPUs, 512 MB to 8 GB RAM.
HighFreq Line
Fixed configurations of high-performance cloud servers with CPU speeds up to 3.6 GHz and memory speeds up to 3,200 MHz.
Suitable for databases such as 1C Bitrix, game servers and other tasks that require high speed processing and response.
Use only with a local boot disk — this helps avoid network latency.
Available from 1 to 8 vCPUs, 2 GB to 64 GB RAM, 30 GB to 960 GB local disk.
Arbitrary configurations
To see the availability of configurations in the regions, see the availability matrix Cloud servers.
An arbitrary configuration can be selected by cloud server creation in the control panel. If you create a cloud server through the OpenStack CLI and Terraform and fixed configurations are not appropriate, create a flavor. Flavors determine the number of vCPUs, RAM, and local disk size (optional) of the server. Through the OpenStack API, you can create a GPU flavor.
Values of arbitrary configurations
In arbitrary configurations, you can select any resource ratio and add GPUs. The available values depend on pool segment.
* If there are more than 8 vCPUs in a configuration, the vCPU:RAM ratio must be at least 1:2. For example, 10 vCPUs require at least 20 GB of RAM.
** If the configuration has more than 8 vCPUs, the vCPU:Local Disk ratio must be at least 1:32. For example, 10 vCPUs require a disk size of at least 320 GB.
If arbitrary configurations are not suitable, you can order your own configuration. Create a ticket and indicate the ratio of resources:
- vCPU:RAM — at least 1:2;
- vCPU:RAM:Local disk — at least 1:2:16.
Create a flavor
To create cloud servers through the OpenStack CLI and Terraform, flavorings are used to define the number of vCPUs, RAM, and local disk size (optional) of the server. Through the OpenStack API, you can create a GPU flavor. Flavors correspond to arbitrary configurations, when creating a flavor consider the available values of arbitrary configurations.
OpenStack CLI
Terraform
OpenStack API
Flavor will only be available for the one project and pool for which you are configured authorization in OpenStack API.
-
Create a flavor:
openstack flavor create \
--private \
--vcpus <vcpu> \
--ram <ram_size> \
--disk <disk_size> \
<flavor_name>Specify:
<vcpu>— number of vCPUs;<ram_size>— RAM size in MB;- optional:
<disk_size>— local disk size in GB. To create a flavor with a network disk, the value must be zero; <flavor_name>— flavor name. It must be unique and not coincide with the names of previously deleted flavors. The list of existing flavors can be viewed usingopenstack flavor list
Example of creating a flavor with 1 vCPU, 1 GB RAM, local disk size of 5 GB:
openstack flavor create --private --vcpus 1 --ram 1024 --disk 5 new_flavor
Use the instructions Create a flavor in the Terraform documentation.
You can only create a GPU flavor through the OpenStack API.
-
Get X-Auth-Token (scope: account) or X-Auth-Token (scope: project) — IAM token for an account or project.
-
In the query output, copy the value
X-Subject-Token. -
Open the CLI on the local computer.
-
Create a request to create a flavor:
curl 'https://<pool>.cloud.api.selcloud.ru/compute/v2.1/flavors' \
-H 'X-Auth-Token: <x-auth-token>' \
-H 'Content-Type: application/json' \
--data-raw '{"flavor":{"name":"<flavor_name>","os-flavor-access:is_public":false,"vcpus":<vcpu_amount>,"ram":<ram_size>,"disk":<disk_size>,"gpu":"<gpu_type>:<gpu_amount>"}}'Specify:
<pool>— pool where the flavor will be created, for example.ru-7. The address (URL) depends on the region and pool, you can look in the URL list;<x-auth-token>— IAM token that you copied in step 2;<flavor_name>— flavor name. It must be unique and not coincide with the names of previously deleted flavors;<vcpu_amount>— number of vCPUs;<ram_size>— RAM size in MB;- optional:
<disk_size>— local disk size in GB. Only NVIDIA® A100 40Gb, NVIDIA® A100 80Gb, or NVIDIA® A30 in the ru-7a and ru-7b pool segment can be used for local disk flavor. To create a network disk flavor, the value must be zero; <gpu_type>— GPU name, e.g.T4,A100(without wordsNVIDIA®,Tesla). The list can be viewed in the table Available GPUs;<gpu_amount>— number of GPUs.
Sample request to create a flavor with 2 vCPUs, 2 GB RAM, bootable network disk, and NVIDIA® Tesla T4 GPU in the ru-7a pool segment:
curl 'https://ru-7.cloud.api.selcloud.ru/compute/v2.1/flavors' \
-H 'X-Auth-Token: AbCD813261b...' \
-H 'Content-Type: application/json' \
--data-raw '{"flavor":{"name":"flavor-name","os-flavor-access:is_public":false,"vcpus":2,"ram":2048,"disk":0,"gpu":"T4:1"}}'
Cloud server flavor list
Flavors correspond to cloud server configurations and define the number of vCPUs, RAM, and local disk size (optional) of the server.
Flavor IDs or names are used to create cloud servers through the OpenStack CLI and Terraform. IDs differ in poolahs.
For example, 4011 — ID, a RAM1.2-16384 — Flavor name that corresponds to a fixed Memory Line configuration with 2 vCPUs, 16 GB RAM in a ru-9 pool.
You can see list of fixed-configuration flavorings in all pools tabulated or to see a list of flavors of all configurations in a certain pool through the OpenStack CLI.
List of fixed-configuration flavorings in all pools
St. Petersburg
Moscow
Novosibirsk
Tashkent
Almaty
Nairobi
ru-3
ru-1
ru-9
ru-2
ru-7
gis-1
ru-8
uz-1
uz-2
kz-1
ke-1
Here:
ID— Cloud Server Flavor ID;Имя— The name of the flavor that corresponds to the fixed configuration ruler:m1.XX— OpenStack basic configurations;SL1.XX— fixed configurations of the Standard Line;CPU1.XX— fixed configurations of the CPU Line;RAM1.XX— Memory Line fixed configurations;GL2.XX— GPU Line fixed configurations;PRC10.XX— Shared Line fixed configurations with vCPU share of 10%;PRC20.XX— Shared Line fixed configurations with vCPU share of 20%;PRC50.XX— Shared Line fixed configurations with vCPU share of 50%;HFL1.XX— fixed configurations of the HighFreq Line;
vCPU— number of vCPUs;RAM, МБ— RAM size in MB;Размер локального диска, ГБ— local disk size in GB.
View a list of all configurations' flavors in a specific pool
Through the OpenStack CLI, you can see a list of flavors that match fixed configurations and flavors that you established.
-
Check out the list of flavors:
openstack flavor listExample answer for pool ru-9 (abbreviated):
+------------+-----------------------+--------+------+-----------+-------+-----------+
| ID | Name | RAM | Disk | Ephemeral | VCPUs | Is Public |
+------------+-----------------------+--------+------+-----------+-------+-----------+
| 1 | m1.tiny | 512 | 0 | 0 | 1 | True |
| 1011 | SL1.1-1024 | 1024 | 0 | 0 | 1 | True |
| 2011 | CPU1.4-8192 | 8192 | 0 | 0 | 4 | True |
| 4011 | RAM1.2-16384 | 16384 | 0 | 0 | 2 | True |
| 3021 | GL2.6-24576-0-1GPU | 24576 | 0 | 0 | 6 | True |
| 9011 | PRC10.1-512 | 512 | 0 | 0 | 1 | True |
| 9021 | PRC20.1-512 | 512 | 0 | 0 | 1 | True |
| 9051 | PRC50.1-512 | 512 | 0 | 0 | 1 | True |
| 8301 | HFL1.1-2048-30 | 2048 | 30 | 0 | 1 | True |
+------------+-----------------------+--------+------+-----------+-------+-----------+Here:
ID— Cloud Server Flavor ID;Name— The name of the flavor that matches the configuration:m1.XX— OpenStack basic configurations;SL1.XX— fixed configurations of the Standard Line;CPU1.XX— fixed configurations of the CPU Line;RAM1.XX— Memory Line fixed configurations;GL2.XX— GPU Line fixed configurations;PRC10.XX— Shared Line fixed configurations with vCPU share of 10%;PRC20.XX— Shared Line fixed configurations with vCPU share of 20%;PRC50.XX— Shared Line fixed configurations with vCPU share of 50%;HFL1.XX— fixed configurations of the HighFreq Line;
RAM— RAM size in MB;Disk— local disk size in GB;VCPUs— number of vCPUs;Is Public— flavor scope:True— public flavoring,False— private flavors.