Yandex Cloud
Search
Contact UsTry it for free
  • Customer Stories
  • Documentation
  • Blog
  • All Services
  • System Status
    • Featured
    • Infrastructure & Network
    • Data Platform
    • Containers
    • Developer tools
    • Serverless
    • Security
    • Monitoring & Resources
    • AI for business
    • Business tools
  • All Solutions
    • By industry
    • By use case
    • Economics and Pricing
    • Security
    • Technical Support
    • Start testing with double trial credits
    • Cloud credits to scale your IT product
    • Gateway to Russia
    • Cloud for Startups
    • Center for Technologies and Society
    • Yandex Cloud Partner program
    • Price calculator
    • Pricing plans
  • Customer Stories
  • Documentation
  • Blog
© 2025 Direct Cursus Technology L.L.C.
Yandex Managed Service for Valkey™
  • Getting started
    • Resource relationships
      • Active host classes
    • Networking in Yandex Managed Service for Valkey™
    • Sharding
    • Backups
    • Replication and fault tolerance
    • Supported clients
    • Memory management
    • Available commands
    • Quotas and limits
    • Storage
    • Maintenance
    • Valkey™ settings
  • Access management
  • Pricing policy
  • Terraform reference
  • Monitoring metrics
  • Audit Trails events
  • Public materials
  • Release notes
  • FAQ
  1. Concepts
  2. Host classes
  3. Active host classes

Valkey™ host classes

Written by
Yandex Cloud
Updated at November 26, 2025

The host class determines the computing power allocated for each host in a cluster. When you change the class for a cluster, all existing hosts change to match it.

Note

The amount of memory allocated to a host also depends on the maxmemory configuration parameter for Valkey™ hosts: the maximum amount of data equals 75% of available memory. For more information, see Memory management.

The host class determines the available disk types:

  • hm1: network-ssd, local-ssd.
  • hm2, hm3: network-ssd, local-ssd, network-ssd-nonreplicated, network-ssd-io-m3.
  • b2, b3: network-ssd.

The storage space available to the host should be at least twice as large as the selected memory size. For more information on Yandex Managed Service for Valkey™ technical and organizational limitations, see Quotas and limits.

Available host classesAvailable host classes

Configuration types:

  • burstable: Configurations with a guaranteed vCPU share under 100%. Those host classes are intended for test loads. We do not recommend using them for production solutions.

    A cluster with this configuration type can only contain one host per cluster or shard.

  • high-memory: Standard configurations for Valkey™.

    A cluster with this configuration type may contain one or more hosts (within the current quota) per cluster or shard. The minimum number of hosts per cluster depends on the selected disk type.

Note

Not available for clusters with hosts residing in the ru-central1-d availability zone:

  • Intel Broadwell platform
  • Local SSD storage if using Intel Cascade Lake

Alert

Starting January 1, 2024, the following existing host classes are deprecated: b1.nano, b1.micro, b1.small, b2.nano, b2.micro, and b2.small. New hosts of these classes cannot be created since June 20, 2023.

Host class name Number of CPUs RAM, GB CPU performance Disk
size, GB
Intel Broadwell
hm1.nano 2 8 100% 16 - 4,096
hm1.micro 2 12 100% 24 - 4,096
hm1.small 2 16 100% 32 - 4,096
hm1.medium 4 24 100% 48 - 4,096
hm1.large 4 32 100% 64 - 4,096
hm1.xlarge 4 48 100% 96 - 4,096
hm1.2xlarge 4 64 100% 128 - 4,096
hm1.3xlarge 4 80 100% 160 - 4,096
hm1.4xlarge 8 96 100% 192 - 4,096
hm1.5xlarge 8 128 100% 256 - 4,096
hm1.6xlarge 8 160 100% 320 - 4,096
hm1.7xlarge 8 192 100% 384 - 4,096
hm1.8xlarge 8 224 100% 448 - 4,096
hm1.9xlarge 8 256 100% 512 - 4,096
Intel Cascade Lake
b2.medium 2 4 50% 8 - 32
hm2.nano 2 8 100% 16 - 8,184
hm2.micro 2 12 100% 24 - 8,184
hm2.small 4 16 100% 32 - 8,184
hm2.medium 4 24 100% 48 - 8,184
hm2.large 4 32 100% 64 - 8,184
hm2.xlarge 4 48 100% 96 - 8,184
hm2.96xlarge 6 96 100% 192 - 8,184
hm2.128xlarge 8 128 100% 256 - 8,184
hm2.160xlarge 10 160 100% 320 - 8,184
hm2.192xlarge 12 192 100% 384 - 8,184
hm2.224xlarge 14 224 100% 448 - 8,184
hm2.256xlarge 16 256 100% 512 - 8,184
hm2.320xlarge 20 320 100% 640 - 8,184
hm2.384xlarge 24 384 100% 768 - 8,184
hm2.448xlarge 28 448 100% 896 - 8,184
hm2.512xlarge 32 512 100% 1024 - 8,184
Intel Ice Lake
b3-c1-m4 2 4 50% 8 - 32
hm3-c2-m8 2 8 100% 16 - 8,184
hm3-c2-m12 2 12 100% 24 - 8,184
hm3-c4-m16 4 16 100% 32 - 8,184
hm3-c4-m24 4 24 100% 48 - 8,184
hm3-c4-m32 4 32 100% 64 - 8,184
hm3-c4-m48 4 48 100% 96 - 8,184
hm3-c6-m60 6 60 100% 120 - 8,184
hm3-c6-m78 6 78 100% 156 - 8,184
hm3-c6-m96 6 96 100% 192 - 8,184
hm3-c8-m128 8 128 100% 256 - 8,184
hm3-c12-m192 12 192 100% 384 - 8,184
hm3-c14-m224 14 224 100% 448 - 8,184
hm3-c16-m256 16 256 100% 512 - 8,184
hm3-c20-m320 20 320 100% 640 - 8,184
AMD Zen 4
m4a-c2-m16 2 16 100% 32 - 8,184
m4a-c4-m32 4 32 100% 64 - 8,184
m4a-c8-m64 8 64 100% 128 - 8,184
m4a-c16-m128 16 128 100% 256 - 8,184
m4a-c32-m256 32 256 100% 512 - 8,184
m4a-c64-m512 64 512 100% 930 - 8,184
m4a-c96-m768 96 768 100% 1,536 - 8,184
m4a-c128-m1024 128 1,024 100% 2,048 - 8,184
m4a-c224-m1792 224 1,792 100% 3,534 - 8,184
AMD Zen 4 HighFreq
m4af-c2-m16 2 16 100% 32 - 8,184
m4af-c4-m32 4 32 100% 64 - 8,184
m4af-c8-m64 8 64 100% 128 - 8,184
m4af-c16-m128 16 128 100% 256 - 8,184
m4af-c32-m256 32 256 100% 512 - 8,184
m4af-c48-m384 48 384 100% 768 - 8,184
m4af-c80-m640 80 640 100% 1,280 - 8,184
hm4af-c2-m20 2 20 100% 40 - 8,184
hm4af-c2-m24 2 24 100% 48 - 8,184
hm4af-c4-m40 4 40 100% 80 - 8,184
hm4af-c4-m48 4 48 100% 93 - 8,184
hm4af-c8-m80 8 80 100% 160 - 8,184
hm4af-c8-m96 8 96 100% 186 - 8,184
hm4af-c16-m160 16 160 100% 279 - 8,184
hm4af-c16-m192 16 192 100% 384 - 8,184
hm4af-c32-m320 32 320 100% 640 - 8,184
hm4af-c32-m384 32 384 100% 768 - 8,184
hm4af-c48-m480 48 480 100% 930 - 8,184
hm4af-c48-m576 48 576 100% 1,152 - 8,184
hm4af-c80-m800 80 800 100% 1,600 - 8,184
hm4af-c80-m960 80 960 100% 1,860 - 8,184

Was the article helpful?

Previous
Resource relationships
Next
Before June 20, 2023
© 2025 Direct Cursus Technology L.L.C.